Ingredient variables in iron gall ink recipes

Tannin source

Botanical species

Part used - gall nuts, bark, leaves, branches
Form – i.e. green/blue galls, dried bark, etc

Prepared acids: tannic C76H52O46 or gallic (HO)3C6H2CO2H

Liquid – water, vinegar, wine
Iron source – iron(II) sulfate, copperas, vitriol → iron(II) sulfate

Some recipes call for mixing the iron with something first, ie vinegar (levigation of vitriol and vinegar?)

Gum source – gum Arabic, Sarcocolla, Tragacanth gum

Process variables

Ingredient ratios
Preparation time – leaving solutions to settle, soak, etc
Preparation of tannin source – i.e. particle size (crushed, powdered, whole, “bruised”), soaking, fermenting, etc

Efficiency of extraction

Heat – heated or not, boiled, slow heat, etc
Reduction – concentrated or not, by how much

Covered/uncovered

Filtration – solution filtered/sifted or not and when during the process
pH of final solution
Effect of age on final sample – spectral changes of same source at different times
Application substrate – paper type, glass slide, ivory

Decisions on fixed variables

The following will be consistent throughout these preliminary experiments:

Gum source: gum Arabic
Iron source: iron(II) sulfate
Application substrate: pure cellulose paper

Variables to test

Preliminary experiments will focus on the effects of:

The tannin source

Gallic/tannic acids vs. various gall nuts

Phase 1 experiments will test Allepo galls and gallic/tannic acid
Phase 2 will consider other gall nut species (Kew Gardens, Botanical Garden in Slovenia), and possibly reconsider other sources (e.g. rhus corriaria)

Liquid

Phase 1 - water
Phase 2 – comparison to vinegar, wine

Process

Heated vs. unheated
Type of heating
Reduction
Procedure time

Aging

Same sample at different ages

Investigation processes

During preparation, pH testing of individual recipe components, at various stages of the recipe, and of final sample

A-D strips
Electrode from Objects Cons?

Raman – experiment with different wavelengths, power, run time, etc

Microfadeometry on Raman test site

FTIR
UV-vis (may be effective for aged samples)
FCIR of finished samples? [while A. Rizzo is away]

Literature recipes

Paper

Recipe

Breakdown

Ratios (liquid:tannin:iron:gum)

*normalized to liquid*

Senvaitiene - Spectroscopic evaluation and characterization of different historical writing inks

Iron-gall ink; 16 ml of the cold distilled water and 1 ml of 10% acetic acid (Lachema) were added to the powdered oak galls (1 g; Kremer Pigmente), iron (II) sulphate (FeSO47H2O) (0.53 g; Fluka) and gum arabic (0.53 g; Kremer Pigmente). After careful grinding the mixture was stored for 8 weeks with intermediate mixing in a slightly covered vessel and filtered. The pH of the final ink solution was equal 1.66.

Liquid

16 ml of the cold distilled water and 1 ml of 10% acetic acid

0.0586

0.0312

Tannin

1g, powdered Aleppo galls

Iron(II) sulphate

0.53g

Gum Arabic

0.53g

Lee - Raman analysis of iron gall inks on parchment

Approximate 1:1 iron(III) complexes were prepared from the combination of iron(II) sulfate with gallic acid, tannic acid or crushed oak galls in deionised water:

1.22 g gallic acid (3,4,5-trihydroxybenzoic acid; Sigma) was dissolved in 150 mL heated deionised water. 2.00 g iron(II) sulfate (FeSO47H2O; Merck) was dissolved in a small amount of deionised water. The two solutions were combined and the reaction mixture left to oxidize with occasional stirring for a week. The mixture was then filtered and the collected black solid washed repeatedly with deionised water. It was initially dried under vacuum and then air dried.
1.22 g tannic acid (Sigma) was boiled in 150 mL deionized water and combined with 2.00 g iron(II) sulfate dissolved in deionised water. A black solid was collected following the procedure in (i).
3.20 g of crushed oak galls (Kremer Pigmente) were boiled in 150 mL deionised water and combined when cooled with 2.80 g iron(II) sulfate dissolved in deionised water. Once again, a black solid was obtained following the procedure in (i).

(i)

Liquid

150 ml deionized water

0.0081

0.0133

Tannin

1.22g gallic acid

(HO)3C6H2CO2H

Sigma

Iron(II) sulphate

2g dissolved in small amount of water

Gum Arabic

(ii)

Liquid

150ml deionized water

0.0081

0.0133

Tannin

1.22g tannic acid

C76H52O46

Sigma

Iron(II) sulphate

2g dissolved in deionized water

Gum Arabic

(iii)

Liquid

150ml deionized water

0.0213

0.0187

Tannin

3.2g crushed Aleppo galls

Iron(II) sulphate

2.8g dissolved in deionized water

Gum Arabic

Bicchieri - Non-destructive spectroscopic investigation on historic Yemenite scriptorial fragments: evidence of different degradation and recipes for iron tannic inks

**Used by Choi for unheated samples**

Four iron tannic inks prepared using commercial extracts of hydrolysable tannins from chestnut (Castanea sativa), sumach (Rhus coriaria), myrobalan (Prunus cerasifera) and valonia (Quercus macrolepis). As a source of condensed tannins, one extract of mangrove (Rhizophora mangle) was used. The inks were prepared by dissolving 0.78 g of each extract in 5.0 ml of deionized water and then adding 0.31 g of gum arabic. Finally 0.47 g of iron(II) sulphate was dissolved in each solution. After 12 h of the decanting phase, the inks were ready for use.

– Two iron gall inks obtained from Aleppo and Italian (Umbria region) oak-gall nuts. The inks were prepared by dissolving 0.78 g of powdered oak-gall nuts in 5.0 ml of deionized water for 24 h. After filtration, the volume was set to 5.0 ml; gum arabic and iron(II) sulphate were then added in the same way and amount used in the preparation of the other tannic inks.

Liquid

5ml deionized water

0.156

0.094

0.062

Galls

0.78g powdered Aleppo galls (sample 2 with Umbria region galls)

Iron(II) sulphate

0.47g

Gum Arabic

0.31g

Sadra Zekrgoo - Methods of Creating, Testing and

Identifying Traditional Black Persian Inks

2 g gum Arabic, 4 g vitriol (iron(II) sulphate),6 g oak galls, 60 mL water.

The galls are broken into four to five pieces, placed in a bowl and then boiled water is added to it. The bowl is then left under sunlight for more of the essence to be extracted from the galls, and left overnight to settle. Afterwards the contents of the bowl are set to boil and filtered through a cloth. In the next stage, vitriol (iron(II) sulphate) and gum Arabic are added to the liquid while stirring; once the vitriol is added to the essence of galls, it starts turning purple. The mixture is left for a couple of hours at which point it turns darker and darker. The newly made ink is then filtered once again, placed in a bottle and shaken.

Liquid

60ml water

0.1

0.0667

0.0333

Galls

6g oak galls

Iron(II) sulphate

Gum Arabic

Historical Recipes

Source

Recipe

Breakdown

Ratios (liquid:tannin:iron:gum)

*normalized to liquid*

Columbia Conservation Lab (Naomi – from irongallink.org)

Dissolve gum arabic in the wine

Combine the gum arabic/wine mixture with Aleppo galls and iron sulfate

Mix until well combined

Strain through cheese cloth

Liquid

120ml white wine

0.2925

0.1751

0.0035

Tannin

35g, powdered Aleppo galls

Iron(II) sulphate

21g

Gum Arabic

0.5g

Thompson – Manuscript Inks (used by Choi)

From Thompson: All parts by weight.

The oak gall infusion, after a few days of soaking was brought to a slow boil and filtered while still warm, before cooking down to approximately half the original volume. The other materials were weighed out and added with stirring. This formed the basic recipe. All of the other inks were variations on the theme. For instance, the oak bark infusion was cooked down from a volume of 3700 ml to 400 ml. The basic recipe above is very acid, in the range of 1.5 - 2.0 pH; the oak bark ink has a pH of 3.5. Eggshells dissolved in the ink raise the pH, but make the ink more gray than black.

Liquid

30 parts water

0.1

0.0667

0.0333

Tannin

3 parts oak galls

Iron(II) sulphate

2 parts copperas

Gum Arabic

1 part

Griffini

Cooked
Squeezed
Soaked
Dry

(1)

Liquid

10 parts water

0.4

0.1

0.4

Tannin

4 parts tannin tree

Iron

1 part vitriol (w/ ? vinegar)

Gum

4 parts gum

(2)

Liquid

5 parts water

0.2

Tannin

1 part tannin tree

Iron

10 parts vitriol in vinegar

Gum

1 part gum

(3)

Liquid

10 parts water

0.1

0.05

Tannin

1 part tannin tree

Iron

½ part vitriol (w/ ? vinegar)

Gum

½ part

(4)

Liquid

½ part vinegar; water as needed

Tannin

1 part

Iron

½ part vitriol

Gum

1 part gum

Recipe comparison

The ratios of recipe components

Initial amounts called for in the recipe (does not consider any loss of ingredients during manufacture – i.e. reduction through heat or age, filtration, etc)

Normalized to 1 part liquid
Tannic acid content – please see section below

Source	Liquid	Tannin	Iron	Gum	Tannic acid	%acid - total	Heat
Senvaitiene	1.0000	0.0586	0.0312	0.0312	0.0410	3.6596	Unheated
Lee 1	1.0000	0.0081	0.0133	0.0000	0.0081	0.7962	Unheated
Lee 2	1.0000	0.0081	0.0133	0.0000	0.0081	0.7962	Unheated
Lee 3	1.0000	0.0213	0.0187	0.0000	0.0149	1.4324	Unheated
Bicchieri	1.0000	0.1560	0.0940	0.0620	0.1092	8.3238	Unheated
Zekrgoo	1.0000	0.1000	0.0667	0.0333	0.0700	5.8333	Heated
Columbia	1.0000	0.2925	0.1751	0.0035	0.2047	13.9167	Unheated
Thompson	1.0000	0.1000	0.0667	0.0333	0.0700	5.8333	Heated
Griffini-cooked	1.0000	0.4000	0.1000	0.4000	0.2800	14.7368	Heated
Griffini-squeezed	1.0000	0.2000	2.0000	0.2000	0.1400	4.1176	Unheated
Griffini-soaked	1.0000	0.1000	0.0500	0.0500	0.0700	5.8333	Unheated
Griffini-dry	1.0000	2.0000	1.0000	2.0000	1.4000	23.3333	Unheated
Tannin	% tannic acid
Aleppo gall	70
Oak gall	--
"Tannin tree"	--
Gallic acid	100
Tannic acid	100
Liquid
Water
Water+acetic acid
White wine

Recipes – unheated vs. heated

Work completed by Choi

Tested various sources of tannin, using two methods

Unheated – following Bicchieri

Galls powdered
Ratio of ingredients:

Liquid	Tannin	Iron	Gum
1	0.156	0.094	0.062

*Normalized to liquid*

Heated – following Thompson

Galls crushed

Liquid	Tannin	Iron	Gum
1	0.1	0.0667	0.0333

*Normalized to liquid*

Completed gallnut samples (from “ink report summary rev 2”):

Ink/ date finished preparing	Description	Amount dissolved (g), volume of distilled water (mL)	Amount gum Arabic, and ferrous sulfate added	Color of mixture before and after ferrous sulfate added
11. Oak galls 1, 7/27/15 (Kremer)	Unheated	0.36, 2.5	0.15g, 0.15g	Light brown, black
12. oak galls 2, 7/27/15 (Kremer)	Heated, filtered through wet strength paper	0.39, 3.9	0.13g, 0.24g	Light brown, black
13. oak galls 3, 7/30/15 (Kremer)	Heated, filtered through wet strength paper	0.36, 3.6 (1.6 mL after heating and filtration)	0.05g, 0.11g	Light brown, grey black
24. oak galls from VA 3, 8/13/15	Unheated, centrifuged	0.39, 7.8	0.14, 0.15	Light brown
25. oak galls from VA 4, 8/13/15	Heated, filtered through wet strength paper	0.39, 7.8		Light brown
26. Oak galls 5 , 8/12/15	heated, filtered through wet strength paper	0.39, 3.9 (1.5 mL after heating)	0.12, 0.12	Light brown, (golden brown after heating), black
29. oak galls 6, 8/26/15	Unheated, filtered through muslin	0.39, 2.5 (0.75 mL after filtering)	0.15, 0.12	Golden brown, black
30. oak galls 7, 8/31/15	Heated for 35 minutes, filtered through muslin	0.39, 3.9 (0.75 after heating)	0.16, 0.18	Golden brown, grey-black

Tannic acid content

In the above mentioned recipes, there are various sources of tannin, each with their own percent content of tannic acid
Highlighted in yellow in the graphs and tables, the tannin source amount has been adjusted to consider the tannic acid content:

Tannin	% tannic acid
Aleppo gall	70
Oak gall	70*
"Tannin tree"	70*
Gallic acid	100
Tannic acid	100

*For the purposes of calculation, “oak gall” and “tannin tree” were assumed to have the same tannic acid content as Aleppo galls, but these sources are not specific and hence could have less or more
Calculation:

%acid – total: Of initial ingredient amounts, the percent amount of tannic acid out of all parts:

Note: “tannin” and not “tannic acid” is considered in the sum of all parts to account for the contribution of other components of the tannin source in the volume of the mixture of all ingredients

Aleppo gall tannic acid contents were estimated using the following sources, and using the value from a recent paper (2006):

Source	Tannic acid (%)	Gallic acid (%)
Zajacz, 2006	70	--
Trimble, pg 57 – citing Guibourt 1845	65	2
Trimble, pg 58 – citing Watson Smith 1869	61.65	1.6
Mitchell, pg 92	62	--

ID	Citation
Zajacz	Ágnes Zajácz, Gyöngyi Gyémánt, Natale Vittori, Lili Kandra, Aleppo tannin: structural analysis and salivary amylase inhibition, Carbohydrate Research, Volume 342, Issue 5, 9 April 2007, Pages 717-723, ISSN 0008-6215, http://dx.doi.org/10.1016/j.carres.2006.12.016 (http://www.sciencedirect.com/science/article/pii/S0008621506005775 )
Trimble	The tannins. A monograph on the history, preparation, … v.1. Trimble, Henry, 1853-1898
Mitchell	Inks: their composition and manufacture, including methods of examination and a full list of British patents, by C. Ainsworth Mitchell, and T. C. Hepworth. With 46 illustrations, including 4 plates (1904)

Mitchell “Inks: their composition and manufacture…”

Discusses in great deal the discrepancies in tannic acid content of inks and in tannin sources – including summary of previous studies and a comprehensive study of their own (pgs 87-92)
Concludes: to make a good iron gall ink, “each part of ferrous sulphate requires 3 parts of pure gallotannic acid. Since, however, the proportion of tannic acid varies in each kind of material employed, the proportion of tannin material must naturally vary correspondingly”

This might also be useful for another source of modern analysis on the galls Quercus infectoria (Aleppo), but a full text does not seem to be available: http://www.ncbi.nlm.nih.gov/pubmed/866492

Initial experimental samples

Recipe amounts per sample

	Liquid (ml)	Tannin (g)	Iron (g)	Gum (g)	Tannic acid (TA)
Recipe amounts (2.5ml water)
Griffini-cooked	2.50	1.00	0.25	1.00	0.70
Griffini-squeezed	2.50	0.50	5.00	0.50	0.35
Griffini-soaked	2.50	0.25	0.13	0.13	0.18
Griffini-dry	2.50	5.00	2.50	5.00	3.50

Bicchieri	2.50	0.39	0.23	0.16	0.27
Thompson	2.50	0.25	0.17	0.08	0.18

Using methodology described by Bicchieri:

	Liquid (ml)	Tannin (g)	Iron (g)	Gum (g)	Tannic acid (TA)
Unheated - Bicchieri
Tannin
Control	2.50	0.39	0.23	0.16	0.00
TA	2.50	0.00	0.23	0.16	0.27
Time soaked
0 hrs	2.50	0.39	0.23	0.16	0.00
12 hrs	2.50	0.39	0.23	0.16	0.00
24 hrs	2.50	0.39	0.23	0.16	0.00
48 hrs	2.50	0.39	0.23	0.16	0.00
168 hrs (1 week)	2.50	0.39	0.23	0.16	0.00
0 hrs TA	2.50	0.00	0.23	0.16	0.27
12 hrs TA	2.50	0.00	0.23	0.16	0.27
24 hrs TA	2.50	0.00	0.23	0.16	0.27
48 hrs TA	2.50	0.00	0.23	0.16	0.27
168 hrs TA	2.50	0.00	0.23	0.16	0.27
Heat (following Thompson)
Heat	2.50	0.39	0.23	0.16	0.00
Heat TA	2.50	0.00	0.23	0.16	0.27
Reduction
0.33	2.50	0.39	0.23	0.16	0.00
1/3 TA	2.50	0.00	0.23	0.16	0.27
0.67	2.50	0.39	0.23	0.16	0.00
2/3 TA	2.50	0.00	0.23	0.16	0.27
Liquid
Vinegar	2.50	0.39	0.23	0.16	0.00
Vinegar TA	2.50	0.00	0.23	0.16	0.27

Using methodology described by Thompson:

	Liquid (ml)		Tannin (g)	Iron (g)			Gum (g)		Tannic acid (TA)
Heated - Thompson
Tannin
Control	2.50	0.25			0.17	0.08		0.00
TA	2.50	0.00			0.17	0.08		0.18
Time soaked
0 hrs	2.50	0.25			0.17	0.08		0.00
12 hrs	2.50	0.25			0.17	0.08		0.00
24 hrs	2.50	0.25			0.17	0.08		0.00
48 hrs	2.50	0.25			0.17	0.08		0.00
168 hrs (1 week)	2.50	0.25			0.17	0.08		0.00
0 hrs TA	2.50	0.00			0.17	0.08		0.18
12 hrs TA	2.50	0.00			0.17	0.08		0.18
24 hrs TA	2.50	0.00			0.17	0.08		0.18
48 hrs TA	2.50	0.00			0.17	0.08		0.18
168 hrs TA	2.50	0.00			0.17	0.08		0.18
Unheated (following Bicchieri)
Unheated	2.50	0.25			0.17	0.08		0.00
Unheated TA	2.50	0.00			0.17	0.08		0.18
Reduction
No reduction	2.50	0.25			0.17	0.08		0.00
No reduction TA	2.50	0.00			0.17	0.08		0.18
1/3	2.50	0.25			0.17	0.08		0.00
1/3 TA	2.50	0.00			0.17	0.08		0.18
2/3	2.50	0.25			0.17	0.08		0.00
2/3 TA	2.50	0.00			0.17	0.08		0.18
Liquid
Vinegar	2.50	0.25			0.17	0.08		0.00
Vinegar TA	2.50	0.00			0.17	0.08		0.18

Testing recipes described in Griffini:

	Liquid (ml)		Tannin (g)	Iron (g)			Gum (g)		Tannic acid (TA)
Griffini
Cooked	2.5	1			0.25	1		0
Cooked TA	2.5	0			0.25	1		0.7
Squeezed	2.5	0.5			5	0.5		0
Squeezed TA	2.5	0			5	0.5		0.35
Soaked	2.5	0.25			0.125	0.125		0
Soaked TA	2.5	0			0.125	0.125		0.175
Dry	2.5	5			2.5	5		0
Dry TA	2.5	0			2.5	5		3.5

Total materials needed:

Vinegar (ml)	Water (ml)	Allepo (g)	Iron (g)	Gum (g)	Tannic acid
15.00	110.00	13.40	24.12	18.18	9.38

References

Bicchieri, Marina, Michela Monti, Giovanna Piantanida, and Armida Sodo. "Non-destructive Spectroscopic Investigation on Historic Yemenite Scriptorial Fragments: Evidence of Different Degradation and Recipes for Iron Tannic Inks." Anal Bioanal Chem Analytical and Bioanalytical Chemistry 405.8 (2013): 2713-721. Web
Griffini, Eugenio. "La Preparazione Degli Inchiostri Hibr E Midad Di Differenti." Scritti per Il Centenario Della Nascita Di Michele Amari 1 (1990): 444-48. Web.
Lee, Alana S., Vincent Otieno-Alego, and Dudley C. Creagh. "Identification of Iron-gall Inks with Near-infrared Raman Microspectroscopy." J. Raman Spectrosc. Journal of Raman Spectroscopy 39.8 (2008): 1079-084. Web.
Senvaitiene, J., A. Beganskiene, and A. Kareiva. "Spectroscopic Evaluation and Characterization of Different Historical Writing Inks." Vibrational Spectroscopy 37.1 (2005): 61-67. Web.
Lee, Alana S., Peter J. Mahon, and Dudley C. Creagh. "Raman Analysis of Iron Gall Inks on Parchment." Vibrational Spectroscopy 41.2 (2006): 170-75. Web.
Thompson, Jack C., and Claes G. Lindblad. Manuscript Inks: Being a Personal Exploration of the Materials and Modes of Production. Portland, Or.: Caber, 1996. Print.