In this research, we performed radiographic experiments exploiting polychromatic phasecontrast im... more In this research, we performed radiographic experiments exploiting polychromatic phasecontrast imaging, a method for contrast enhancement in transmitted neutron beams by measuring a sample made with a homogeneous material but exhibiting a different microstructure induced by mechanical actions. We were able to reveal the armourers' marks now obscured by polishing that are present in these examples of medieval armour.
Arms and armour were frequently blued by heating to form a surface layer of oxide. This was studi... more Arms and armour were frequently blued by heating to form a surface layer of oxide. This was studied on specimens of armour and swords by the technique of ellipsometry. An elliptically polarised beam of laser light was shone onto the surface and compared with the reflected beam. The layer of blueing was found to consist of two layers, of different oxides, the proportions of which varied with the % of oxygen in the environment during heating. It is suggested that medieval craftsmen could have controlled their blueing by using an enclosed furnace to reduce the proportion of oxygen to carbon monoxide in contact with the heated steel.
In this research, we performed radiographic experiments exploiting polychromatic phasecontrast im... more In this research, we performed radiographic experiments exploiting polychromatic phasecontrast imaging, a method for contrast enhancement in transmitted neutron beams by measuring a sample made with a homogeneous material but exhibiting a different microstructure induced by mechanical actions. We were able to reveal the armourers' marks now obscured by polishing that are present in these examples of medieval armour.
A. Williams, D. Edge 2015, Three 'grotesque' helmet visors in the Wallace Collection, AMM XI: 225... more A. Williams, D. Edge 2015, Three 'grotesque' helmet visors in the Wallace Collection, AMM XI: 225-233 Three 'grotesque' helmet visors in the Wallace Collection have been analysed. Two were made of hardened steel, and are likely to th have been products of the Innsbruck Hofplattnerei (Court Armoury) while the third one is a 19 c. fake. th
There is an ongoing program of research in the Conservation Department of the Wallace Collection ... more There is an ongoing program of research in the Conservation Department of the Wallace Collection into the composition and manufacture of arms and armour. Iron & steel artefacts in Europe were generally made from bloomery metal; that is the iron of steel was never melted, and thus contained appreciable quantities of slag. The carbon content of steel seldom exceeding 0.6% although with suitable heat treatment (quenching & tempering) such a steel could combine hardness with toughness. Those from India (or rather the Indo-Persian area) pose unique problems. In the Indo-Persian area, as well as bloomery metal, a quantity of steel was also made by the crucible process, which yielded small ingots of cast steel. The carbon content might be anywhere between 1.2% and 1.8% and considerable hardness (and very low slag contents) was achieved without heat-treatment. In some cases, these ingots were allowed to cool very slowly within the crucible, and subsequently forged carefully, so that the large crystals of cementite (iron carbide, Fe 3 C) that had formed, were broken up but preserved in rows. After polishing and etching, the surface would show a pattern resembling watered silk ("Damascus steel"). The most highly prized swords and armour were made of such a steel, and a further class (as yet an unknown proportion) were made of crucible steel which did not exhibit a pattern. Our research on broken swords from the Arsenal of Hyderabad 1 has shown that while some Indian swords show a pattern many more were made of crucible steel without such a pattern, and some of these were (even further) effectively hardened by heat-treatment. The cheapest class of swords and armour were simply made of bloomery (low-carbon) steel. What is surprising is that some examples of armour seem to have been made of both types of metal (crucible steel and bloomer iron) joined together. Of course, the Indian arms and armour in the Wallace Collection are not broken, and so their analysis poses diffi cult questions. We are trying to devise methods of employing neutron diffraction to study them non-invasively. The high carbon contents of crucible steels lend themselves to this method of investigation. But there is a further question which the results of analysis raise. Many swords with "Damascus steel" patterns
Edwin Wood, David Edge, Alan Williams 2013, A note on the construction and metallurgy of mail arm... more Edwin Wood, David Edge, Alan Williams 2013, A note on the construction and metallurgy of mail armour exhibited in The Wallace Collection, AMM IX: 203-229 An exhibition was mounted in 2009 at the Wallace Collection, London, displaying a number of mail shirts and other garments. The number of rings used in several of these mail shirts has been accurately determined for the first time. Mail can be "tailored" that is, its shape can be altered, by varying the number of links attached to each one from 4 to 3 or 5, thus contracting or expanding the weave. The "tailoring" of several shirts has been mapped, and the thickness of the links in different parts of the garments compared. Selected links have been studied by metallography, and deductions made about the heat-treatment of the completed shirt.
The first metal to be used for tools and weapons was copper, partly because copper ores are gener... more The first metal to be used for tools and weapons was copper, partly because copper ores are generally brightly coloured minerals which would be attractive as applied decoration, first for bodies, and then later for ceramics, and partly because copper ores are easy to reduce to the metal. Unlike those of copper, iron ores are very widespread, but the extraction of iron is not so simple, because its melting-point is much higher (iron 1550ºC; copper 1080ºC). Any attempt to reduce (or "smelt") iron ores in a simple copper-smelting furnace will give an unusable mixture of iron and slag. Even if the iron ore is of exceptional purity, and contains no earthy matter itself, there is generally sufficient silica (silicon dioxide, SiO 2) present in the stones and clay which make up the wall of the hearth to react with part of the iron ore and form a slag. The iron ore may be treated as iron oxide only. 2 FeO + SiO 2 F 0 E 0 = Fe 2 SiO 4 (iron oxide + silicon oxide F 0
Prva kovina, ki so jo uporabljali za izdelavo orodij in orožij, je bil baker. Deloma zato, ker ba... more Prva kovina, ki so jo uporabljali za izdelavo orodij in orožij, je bil baker. Deloma zato, ker bakrovo rudo po navadi najdemo v obliki privlačnih, svetlo obarvanih mi-neralov. Z njimi so si ljudje krasili telesa in pozneje tudi keramiko. Po drugi strani pa zato, ker bakrovo rudo zlahka stalimo v kovino. Za razliko od bakra je železova ruda zelo razširjena, vendar taljenje železa ni tako preprosto, kajti tališče je veliko višje (železo 1550º C; baker 1080º C). Kakršen koli poskus, da bi železovo rudo stalili v preprosti peči za taljenje bakra, bi prinesel le neuporabno mešanico železa in žlindre. Četudi je železova ruda izjemno čista in ne vsebuje primesi prsti, je po navadi v kamnih in glini dovolj kremena (silicijevega dioksida SiO 2), da stena ognjišča reagira z deležem železove rude in ustvari žlindro. V osnovi lahko železovo rudo opišemo kot železov oksid. 2 FeO + SiO 2 Fe 2 SiO 4 (železov oksid + silicijev oksid železov silikat) Čeprav bi lahko redukcija železove rude potekala že pri 700–800º C, mora temperatura v talilni peči doseči vsaj 1200º C, da se žlindra stopi in loči od železa. Ko se žlindra utekočini, se loči od še vedno trdnega železa, ki se oblikuje v kepo (ali »volk«). Ta je porozen in vsebuje zelo malo raztopljenega ogljika, a veliko ujete žlindre. Takšne peči so zaradi tega znane kot peči na volk, njihov proizvod pa je ma-loogljično jeklo. Da utrdimo volk in iz njega odstranimo večino žlindre, sta potrebni večkratno segrevanje in kovanje. Maloogljično jeklo iz takšnih peči je bilo osnovni material za izdelovanje orožja in oklepov na območju Evrope v srednjem veku.
Wallace Collection A20 is a composite armour showing components made of such different metal that... more Wallace Collection A20 is a composite armour showing components made of such different metal that it was assembled from the products of at least five workshops.
Neutron Diffraction represents the ideal technique for the characterisation of the micro-structur... more Neutron Diffraction represents the ideal technique for the characterisation of the micro-structural properties of ancient metals, allowing retrieval of information on the smelting process, and the mechanical and thermal treatments applied during the manufacture of the sample.
In this research, we performed radiographic experiments exploiting polychromatic phasecontrast im... more In this research, we performed radiographic experiments exploiting polychromatic phasecontrast imaging, a method for contrast enhancement in transmitted neutron beams by measuring a sample made with a homogeneous material but exhibiting a different microstructure induced by mechanical actions. We were able to reveal the armourers' marks now obscured by polishing that are present in these examples of medieval armour.
Arms and armour were frequently blued by heating to form a surface layer of oxide. This was studi... more Arms and armour were frequently blued by heating to form a surface layer of oxide. This was studied on specimens of armour and swords by the technique of ellipsometry. An elliptically polarised beam of laser light was shone onto the surface and compared with the reflected beam. The layer of blueing was found to consist of two layers, of different oxides, the proportions of which varied with the % of oxygen in the environment during heating. It is suggested that medieval craftsmen could have controlled their blueing by using an enclosed furnace to reduce the proportion of oxygen to carbon monoxide in contact with the heated steel.
In this research, we performed radiographic experiments exploiting polychromatic phasecontrast im... more In this research, we performed radiographic experiments exploiting polychromatic phasecontrast imaging, a method for contrast enhancement in transmitted neutron beams by measuring a sample made with a homogeneous material but exhibiting a different microstructure induced by mechanical actions. We were able to reveal the armourers' marks now obscured by polishing that are present in these examples of medieval armour.
A. Williams, D. Edge 2015, Three 'grotesque' helmet visors in the Wallace Collection, AMM XI: 225... more A. Williams, D. Edge 2015, Three 'grotesque' helmet visors in the Wallace Collection, AMM XI: 225-233 Three 'grotesque' helmet visors in the Wallace Collection have been analysed. Two were made of hardened steel, and are likely to th have been products of the Innsbruck Hofplattnerei (Court Armoury) while the third one is a 19 c. fake. th
There is an ongoing program of research in the Conservation Department of the Wallace Collection ... more There is an ongoing program of research in the Conservation Department of the Wallace Collection into the composition and manufacture of arms and armour. Iron & steel artefacts in Europe were generally made from bloomery metal; that is the iron of steel was never melted, and thus contained appreciable quantities of slag. The carbon content of steel seldom exceeding 0.6% although with suitable heat treatment (quenching & tempering) such a steel could combine hardness with toughness. Those from India (or rather the Indo-Persian area) pose unique problems. In the Indo-Persian area, as well as bloomery metal, a quantity of steel was also made by the crucible process, which yielded small ingots of cast steel. The carbon content might be anywhere between 1.2% and 1.8% and considerable hardness (and very low slag contents) was achieved without heat-treatment. In some cases, these ingots were allowed to cool very slowly within the crucible, and subsequently forged carefully, so that the large crystals of cementite (iron carbide, Fe 3 C) that had formed, were broken up but preserved in rows. After polishing and etching, the surface would show a pattern resembling watered silk ("Damascus steel"). The most highly prized swords and armour were made of such a steel, and a further class (as yet an unknown proportion) were made of crucible steel which did not exhibit a pattern. Our research on broken swords from the Arsenal of Hyderabad 1 has shown that while some Indian swords show a pattern many more were made of crucible steel without such a pattern, and some of these were (even further) effectively hardened by heat-treatment. The cheapest class of swords and armour were simply made of bloomery (low-carbon) steel. What is surprising is that some examples of armour seem to have been made of both types of metal (crucible steel and bloomer iron) joined together. Of course, the Indian arms and armour in the Wallace Collection are not broken, and so their analysis poses diffi cult questions. We are trying to devise methods of employing neutron diffraction to study them non-invasively. The high carbon contents of crucible steels lend themselves to this method of investigation. But there is a further question which the results of analysis raise. Many swords with "Damascus steel" patterns
Edwin Wood, David Edge, Alan Williams 2013, A note on the construction and metallurgy of mail arm... more Edwin Wood, David Edge, Alan Williams 2013, A note on the construction and metallurgy of mail armour exhibited in The Wallace Collection, AMM IX: 203-229 An exhibition was mounted in 2009 at the Wallace Collection, London, displaying a number of mail shirts and other garments. The number of rings used in several of these mail shirts has been accurately determined for the first time. Mail can be "tailored" that is, its shape can be altered, by varying the number of links attached to each one from 4 to 3 or 5, thus contracting or expanding the weave. The "tailoring" of several shirts has been mapped, and the thickness of the links in different parts of the garments compared. Selected links have been studied by metallography, and deductions made about the heat-treatment of the completed shirt.
The first metal to be used for tools and weapons was copper, partly because copper ores are gener... more The first metal to be used for tools and weapons was copper, partly because copper ores are generally brightly coloured minerals which would be attractive as applied decoration, first for bodies, and then later for ceramics, and partly because copper ores are easy to reduce to the metal. Unlike those of copper, iron ores are very widespread, but the extraction of iron is not so simple, because its melting-point is much higher (iron 1550ºC; copper 1080ºC). Any attempt to reduce (or "smelt") iron ores in a simple copper-smelting furnace will give an unusable mixture of iron and slag. Even if the iron ore is of exceptional purity, and contains no earthy matter itself, there is generally sufficient silica (silicon dioxide, SiO 2) present in the stones and clay which make up the wall of the hearth to react with part of the iron ore and form a slag. The iron ore may be treated as iron oxide only. 2 FeO + SiO 2 F 0 E 0 = Fe 2 SiO 4 (iron oxide + silicon oxide F 0
Prva kovina, ki so jo uporabljali za izdelavo orodij in orožij, je bil baker. Deloma zato, ker ba... more Prva kovina, ki so jo uporabljali za izdelavo orodij in orožij, je bil baker. Deloma zato, ker bakrovo rudo po navadi najdemo v obliki privlačnih, svetlo obarvanih mi-neralov. Z njimi so si ljudje krasili telesa in pozneje tudi keramiko. Po drugi strani pa zato, ker bakrovo rudo zlahka stalimo v kovino. Za razliko od bakra je železova ruda zelo razširjena, vendar taljenje železa ni tako preprosto, kajti tališče je veliko višje (železo 1550º C; baker 1080º C). Kakršen koli poskus, da bi železovo rudo stalili v preprosti peči za taljenje bakra, bi prinesel le neuporabno mešanico železa in žlindre. Četudi je železova ruda izjemno čista in ne vsebuje primesi prsti, je po navadi v kamnih in glini dovolj kremena (silicijevega dioksida SiO 2), da stena ognjišča reagira z deležem železove rude in ustvari žlindro. V osnovi lahko železovo rudo opišemo kot železov oksid. 2 FeO + SiO 2 Fe 2 SiO 4 (železov oksid + silicijev oksid železov silikat) Čeprav bi lahko redukcija železove rude potekala že pri 700–800º C, mora temperatura v talilni peči doseči vsaj 1200º C, da se žlindra stopi in loči od železa. Ko se žlindra utekočini, se loči od še vedno trdnega železa, ki se oblikuje v kepo (ali »volk«). Ta je porozen in vsebuje zelo malo raztopljenega ogljika, a veliko ujete žlindre. Takšne peči so zaradi tega znane kot peči na volk, njihov proizvod pa je ma-loogljično jeklo. Da utrdimo volk in iz njega odstranimo večino žlindre, sta potrebni večkratno segrevanje in kovanje. Maloogljično jeklo iz takšnih peči je bilo osnovni material za izdelovanje orožja in oklepov na območju Evrope v srednjem veku.
Wallace Collection A20 is a composite armour showing components made of such different metal that... more Wallace Collection A20 is a composite armour showing components made of such different metal that it was assembled from the products of at least five workshops.
Neutron Diffraction represents the ideal technique for the characterisation of the micro-structur... more Neutron Diffraction represents the ideal technique for the characterisation of the micro-structural properties of ancient metals, allowing retrieval of information on the smelting process, and the mechanical and thermal treatments applied during the manufacture of the sample.
The effectiveness of the lance in mounted warfare and tournaments has been investigated by severa... more The effectiveness of the lance in mounted warfare and tournaments has been investigated by several riders attacking a target (of known weight) arranged to function as a ballistic pendulum. The height to which the target rose was measured by video recording and so the impact energy of the lance under different circumstances has been calculated. The results showed that while the use of stirrups showed some increase in energy, the use of the lance arret (rest) and rigid breast plate showed a greater increase. The maximum energies were limited by the lance breaking and were up to around 250J.
Bullets dents in many examples of 17th century armour do not correspond to damage inflicted in ba... more Bullets dents in many examples of 17th century armour do not correspond to damage inflicted in battle, as impact energy calculations on armour worn in battle demonstrates.
ABSTRACT We present novel results from a non-invasive examination of two kabuto (helmets), made i... more ABSTRACT We present novel results from a non-invasive examination of two kabuto (helmets), made in Japan in the 17th century. Neutron-imaging experiments (radiography and tomography), carried out at the ICON and NEUTRA beamlines, operating at the neutron source SINQ (CH), have allowed to determine the inner metal structure and manufacturing techniques of these beautiful examples of past technology, revealing some otherwise invisible details.
Neutron time-of-flight diffraction technique has been used to characterize some Japanese historic... more Neutron time-of-flight diffraction technique has been used to characterize some Japanese historical artifacts. With this method, metal samples can be analyzed in their bulk properties without need of sampling. Results shown here were obtained at the Italian Neutron Experimental Station (INES@ISIS) located at the pulsed neutron source ISIS (UK). The parallel use of a scanning electron microscope equipped with an energy-dispersive X-ray fluorescence device (SEM-EDX) permitted a full quantitative characterization of the investigated samples, namely four hand-guards (Tsubas) of Japanese swords attributed to the Tokugawa age. In particular, we were able to obtain, in a totally non-invasive non-destructive way, a full quantitative phase characterization of the samples, a detailed Bragg peak broadening analysis, and a quantitative texture determination. These results, complemented with those obtained via the traditional analysis method of SEM-EDX, allowed a full characterization of both the bulk and the surface of the artifacts.
The analyses of two helmets and a spatha (probably from the 5th century) are given. The metallur... more The analyses of two helmets and a spatha (probably from the 5th century) are given. The metallurgy of Roman armour in general is also discussed.
A gun of variable barrel length was used with gunpowder of different compositions to try and asse... more A gun of variable barrel length was used with gunpowder of different compositions to try and assess the effectiveness of 15th century handguns.
One notable conclusion was that the effectiveness of gunpowder depended upon the wet-mixing of the ingredients.
Bullet dents which have been taken as an indication that an armour was "bulletproof" are frequen... more Bullet dents which have been taken as an indication that an armour was "bulletproof" are frequently nothing of the sort.
Their size does not not correspond to impacts generated by the likely impact energies of bullets at short range, and so one must conclude that many were fraudulent in intent !
Cast iron was better known in Medieval Europe than has been supposed in the past. There is manu... more Cast iron was better known in Medieval Europe than has been supposed in the past. There is manuscript evidence as well as archaeometallurgical evidence that cast iron, high-carbon steels (and other ferrous alloys) were known to medieval scholars.
An experimental attempt to duplicate the medieval production of saltpetre (potassium nitrate, ess... more An experimental attempt to duplicate the medieval production of saltpetre (potassium nitrate, essential for gunpowder) by means of a "nitre-bed".
At the end of the Middle-Ages, armor is an object both for everyday military use and a luxury att... more At the end of the Middle-Ages, armor is an object both for everyday military use and a luxury attire produced in prestigious European centers of production like Milan located in the north of Italy and Nuremberg or Augsburg in the south of Germany. Even if some research works have focused on that subject nevertheless several questions concerning the nature of the material employed, the technical skills of the craftsmen (quenching and tempering, presence of second phase particles, etc), depending on their uses and their location are still unanswered. Besides, at the same period, crucial changes occurred in the iron and ferrous alloy production chain with the emergence of the new indirect process, the blast furnace and finery. It allows producing great quantities of metal at low cost. Former studies have already showed specific skills for instance in the hardening of steel or in material choice depending on the center of production(Williams, 2003). However a linkage between the kind of reduction process used to obtain the metal and the mass production of armor in military use is a question left unresolved. Furthermore, while centers of production like Milan or Nuremberg are clearly identified by historical sources, very little is known about the origins of the metal employed. A first set of armor coming from Italy Germany and France were examined following the methodology developed in NIMBE/LAPA laboratory (metallography, MEB-EDS and LA-ICP-MS analyses) (Dillmann and L'Héritier, 2007), (Disser et al., 2014). However armor’s samples contain small size of inclusions (diameter <30μm), compared to other iron artifacts (diameter >80μm). Thus a specific methodology has been developed in order to quantify trace elements. Two different equipments have been used: an inductively coupled plasma mass spectrometer, coupled with a nanosecond laser at the CEB center in Orleans, and with a femtosecond laser at IPREM laboratory in Pau. Primary results show the nature and quality of the metal employed, and the kind of reduction process used depending on their location. Furthermore it also allows making a first comparison with existing geochemical databases in order to test hypotheses on the origin of the materials employed
Uploads
Books by Alan Williams
Papers by Alan Williams
One notable conclusion was that the effectiveness of gunpowder depended upon the wet-mixing of the ingredients.
Their size does not not correspond to impacts generated by the likely impact energies of bullets at short range, and so one must conclude that many were fraudulent in intent !