Ancient Monuments Laboratory
Report 66/86
THE MANUFACTURING TECHNIQUES USED
IN ANGLO-SCANDINAVIAN IRON
ARTIFACTS FROM 19-22 COPPERGATE
YORK.
J G McDonnell BTech MIFA
AML reports are interim reports which make available the results
of specialist investigations in advance of full publication.
They are not subject to external refereeing, and their
conclusions may sometimes have to be modified in the light
of archaeological information that was not available at the time
of the investigation. Readers are therefore asked to consult
the author before citing the report in any publication and to
consult the final excavation report when available.
Opinions expressed in AML reports are those of the author and
are not necessarily those of the Historic Buildings and
Monuments Commission for England.
Ancient Monuments Laboratory Report
66/86
THE MANUFACTURING TECHNIQUES USED
IN ANGLO-SCANDINAVIAN IRON
ARTIFACTS FROM 19-22 COPPERGATE
YORK.
J G McDonnell BTech MIFA
Summary
This report summarises the metallographic examination
carried out on the iron objects from the Coppergate
excavations & discusses the relationships between
structure and typology for 50 knives.
Author's address :Department of Mechanical & Production Engineering
University of Aston
Aston Triangle
Birmingham
B4 7ET
021 359 3611 x4314
©
:-Iistori-.::: 3uil--ii~'1gs J.:l.d ~4onuments ~ommission for Sngland
H.B.M.C.E. FUNDED ARCHAEOMETALLURGY CONTRACT
DEPARTMENT OF MECHANICAL AND PRODUCTION ENGINEERING
ASTON UNIVERSITY,
ASTON TRIANGLE
BIRMINGHAM B4 7ET
THE
MANUFACTURING
TECHNIQUES
USED IN
ANGLO-SCANDINAVIAN
IRON
ARTEFACTS FROM 19-22 COPPERGATE, YORK.
By J.G. McDonnell B.Tech. M.I.F.A.
The
manufacturing
Firstly,
stage process.
Secondly,
two
the ore was smelted to produce a bloom
the bloom was refined and subsequently
re-
into bar-iron and finally into artefacts by means of
the
of iron.
worked
cycle of iron ore to artefact was a
The products of the manufacturing
smithing process.
the smelting residues,
1),
Closer
examination
understanding
of
the smithing residues, and the artefacts,
all of which contain evidence about the processes.
(Figure
Analysis
cycle were
to
of
be
these
products
gained of
early
enables
ironworking
the artefacts is the best source
greater
a
of
technology.
technological
data, the composition of the iron and some of the slag inclusions
relates
to
the
smelting
process,
while
the
metallographic
information details the techniques used in the smithing process.
The
from
analysis
Coppergate,
of one hundred artefacts of Viking
York,
afforded an opportunity to
skills of Anglo-Scandinavian blacksmiths.
Age
Date
assess
the
The range of artefacts
examined is shown in Table l, the evidence can be summarised by a
description
examined.
firstly,
considered
of
These
the
are
results
of
obtained
from
special interest
the
for
fifty
three
knives
reasons;
the knives were a common artefact, some of which can be
to have specialist functions.
1
Secondly,
the knives
were
analysed
typologically
Archaeological
Trust)
manufacturing
typology
and
investigated.
Thirdly,
a
(by Mr
P.
and therefore,
Ottaway
of
the
the relationship
archaeological
significant
between
typology
number of
York
can
knives
be
were
available for analysis.
The
terminology used in describing the knives is
Figure 2.
as
1n
There were five possible methods of knife manufacture
shown in Figure 3;
had
shown
Type 0 had no steel cutting edge.
a central steel strip.
Type 2 had the steel
butt (or scarf) welded to the knife back.
Type 1
cutting
edge
Type 3 was all steel.
Type 4 had the cutting edge welded around the core, or was formed
as a result of carburisation of the surface of the iron core.
All
the
artefacts
and
photographed
selected
X-radiographed,
for
analysis
the
latter
were
being
drawn,
repeated
several times under varying conditions.
This enabled details of
the
determined,
artefacts'
indicated
macro-structure
areas
of
corrosion
to
be
which were to
be
and
also
avoided
when
In a recent study of Anglo-Saxon knives from Hamwih
sectioning.
(Southampton) it became possible to determine whether a knife was
of Type 2 (butt welded) manufacture by examination of radiographs
obtained in the course of routine post-excavation screening.
radiographic
evidence
refutes the theory,
and subsequent metallographic
work
The
also
common among many workers in this field that
steel corrodes preferentially to ferritic/phosphoric iron.
A single or two half-sections (Figure 2) were cut from each knife
using the spark-erosion method,
usual manner.
Types
and
the
and prepared and examined in the
Each knife was ascribed to one of the Manufacturing
results,
correlated
Typology are shown in Table 2.
2
with
the
Archaeological
TABLE 1 COPPERGATE ARTEFACT TYPES ANALYSED
ARTEFACT TYPE
NUMBER
ANALYSED
EDGED TOOLS
ARROWHEAD
1
ARROWHEAD POINT?
1
AXE
1
FILE
1
KNIVES
50
PUNCHES
4
SEWING NEEDLES
3
SHEARS
1
SPOKESHAVE
1
SPOON BIT
1
SWORD FRAGMENT
1
WEDGE
1
66
OTHER ARTEFACTS
ANVIL
1
BAR IRON
1
FERRULE
1
HOOK
1
HOOK AND EYE
2
KEY
1
ROD SHAPED BLANKS
24
STAPLE
2
STRAP END
1
34
100
3
TABLE 2
SUMMARY OF VIKING AGE KNIVES FROM COPPERGATE, YORK
Manufacturing Typology
Archaeological
Typology
0
A
B
1
c
2
1
2
3
1
7
1
9
1
2
6
10
D
3
2
2
E2
1
2
1
E3
7
El
4
Specialist Knives
1
Unascribed Knives
1
Totals
9
The
Type
0
2
either
the
1
1
2
13
4
7
1
2
4
1
19
13
3
6
50
knives are a problem since it is
significant
Type
Total
4
number
knives
probable
could have been originally
and have subsequently lost
through wear or corrosion.
that
manufactured
their
cutting
would
not
have been manufactured in any
Therefore,
significant
as
edge
The overall good quality
knives does suggest that knives without steel cutting
a
of
edges
quantity.
is reasonable to group the Type 0 knives with the
it
Type 2's.
Table 2 shows that the majority of knives were
according
that
to
the Type l or the Type 2 methods.
the Typological Group A knives (the
backed
knives),
manufactured
It also
characteristic
shows
angle-
were predominantly of Type 2 manufacture.
The
metallographic evidence and Vicker's Hardness values showed
that
the
were
butt
welded knives Type 2 (mean
consistently
of
better
quality
4
H.V.:717,
than
the
5.0.:177)
other
methods
of
manufacture.
The Type 1 knives for example,
value of H.V.=467, S.D.=230.
had a mean hardness
There were knives of poorer quality
but these probably resulted from deliberate post-manufacture heat
treatments,
e.g.
over tempering,
or accidental tempering in
a
hearth or fire or corrosion of the effective cutting edge.
of
evidence
The
the
metallurgical
study
clearly
demonstrates that the Anglo-Scandinavian smiths had four types of
'iron' available for use; ferritic iron containing less than 0.5%
alloying
elements,
alloying
element,
element.
The
phosphoric iron containing phosphorus as
and
steel containing carbon as
of two or three of the basic
irons.
iron
the
the
alloying
alloying
fourth type of iron used was a banded or
composite
and
an
steel
element
varied widely in
present,
both
The
piled
phosphoric
percentage
within artefacts
an
and
of
the
between
artefacts.
Ferritic Iron
The quantity of iron used that could have been derived
strips
of ferrite iron was very low.
from
It mostly occurred as one
component in banded or piled structures.
Metallographically
it
hardness varied considerably,
of
the
easily
recognisable,
very low carbon steels,
and
but was rarely below lOOHV ..
ferrite appeared to be 'dirty',
presence of fine etch pits.
of
is
which was
due
to
its
Some
the
This is indicative of quench ageing
e.g.
D.03%C.
The presence of such
metallographic structure would normally be included as a ferritic
structure.
5
Phosphoric Iron
Phosphoric iron was used in the manufacture of the
of knife backs,
majority
and was also used in all the structural ironwork
analysed.
The majority of British iron ores contain phosphorus; in the
reduction process it is expected to be concentrated in the metal.
The
effect
inhibits
steel.
of
phosphorus in
carbon
Secondly,
It is,
iron.
diffusion,
metallurgy,
it
is
two-fold;
firstly,
it
thus preventing the manufacture
of
increases the hardness,
therefore,
might
but
iron
embrittling
the
regarded as detrimental in modern day
be construed as
advantageous
in
early
times.
The
presence
metallographically
'ghost phases',
increased
of
phosphoric
in three ways;
secondly,
grain
size.
iron
can
firstly,
be
identified
by the presence
of
by increased hardness and thirdly, by
The ghost phases contained the
maximum
phosphorus content measured (0. 5~•).
Steel
Steel
was used to manufacture the cutting edges of all
knives in which the cutting edge survived,
Type D.
All
steel,
between
showed
the
i.e. all Types except
It was also present in some of the knife backs.
the
structures
although
samples.
that
the
carbon
The
indicated the
use
contents did vary
metallographic
increased
hardness
of
hypo-eutectoid
within
analysis of
was
obtained
samples
and
the
knives
by
slack
quenching, tempered martensites and bainites being predominant.
6
Piled and Banded Structures
The
term piled or banded structure refers to metallographic
structures
in
which
alternate bands
of
differing
1rons
are
present.
The thickness of the bands may vary from a single grain
upwards,
and
the
number of bands also varies.
The bands
can
comprise:
It
is
ferritic iron
I
phosphoric iron
ferritic iron
I
steel
phosphoric iron
I
steel
not
structures
possible
to determine
whether
not
were the result of deliberate manufacture,
welding together of strips of different irons.
may
or
the
piled
i.e.
the
These structures
also have been accidentally generated by the segregation
either
phopshorus or carbon during the smelting process and
subsequent refining of the bloom.
of
the
The latter process would have
comprised numerous re-heating and hammering processes which would
have encouraged segregation.
Pattern-welding
structure.
were
In
can be considered a coarser form
this
structures
coarser,
piled
process strips of differing types of
welded and folded together.
piled
of
in two ways;
iron
Pattern-welding differs
firstly,
the banding
from
was
much
and secondly, the finished product was intended to form
a decorative design.
Two pattern-welded knives were identified,
one of which was sectioned, and shown to be of Type 2 manufacture
with a butt welded steel cutting edge.
in
The pattern was developed
the knife back and comprised thin strips of steel
welded
to
ferritic iron rods.
The
specific
four types of iron were carefully selected and used for
purposes
in the manufacture of the
7
knives
and
other
artefacts.
the
Phosphoric iron was the most common material,
maJor
greatest
other
component
of the knife back,
proportion of the knife.
artefacts,
in particular,
phosphoric
iron
was
the
itself
the
It was also used in all
the
it was the iron most
manufacture the structural ironwork.
that
which was
being
used
to
It can therefore, be stated
commonest
type
of
iron,
and
therefore, probably the cheapest.
Piled
structures,
of bands,
number
of varying quality,
thickness and
was also common in the knives but
used in the structural ironwork.
have
i.e.
was
rarely
It would appear, therefore, to
been retained for use in edged tools,
perhaps as a
poorer
form of steel.
Steel
unless
was specifically used for the cutting edges of tools,
it
was either incorporated in a piled structure
or
the
whole artefact was manufactured from steel.
Ferritic iron was present in piled structures but was rarely
used
there
as an individual component within a
are
For
example,
no knife backs manufactured only from ferritic
In this sense it
The
knife.
model
Scandinavian
iron.
was the rarest form of iron used in artefacts.
proposed
York
for
artefact
manufacture
envisaged either bloom iron or
different
types
being
traded
fabricated
into
artefacts
in
into
York
specialist
in
bar
and
smiths
Angloiron
of
subsequently
shops.
The
presence of smithing slag and rod shaped blanks on the Coppergate
Site indicated the presence, either on the site or very close by,
of such a smithy.
The analysis of the rod shaped blanks
showed
them to be mostly phosphoric or piled irons.
The
source(s)
of
the different types of
8
iron
cannot
be
determined,
but
fundamental
some
general
comments
may
be
made.
The
question is whether the phosphoric iron and ferritic
1ron and steel derived from different ore sources?
The tendency
has
ore
been
to
containing
common
the
ascribe steel and ferritic
low phosphorus contents,
phosphoric ores.
steel
has
phosphoric
iron
but
removed
phosphorus
high
(>5%)
phosphorus
the
method
the smelting furnace
by
may
which
have
steel
was
preferentially
or concentrated it into
the
slag.
observation that has implications on the genesis
steel
the
The analysis of the slag inclusions in
ores,
in
second
sources
Its presence may, therefore indicate use of
manufactured
the
to
and phosphoric iron to
shown that some contain
pentoxide contents.
iron
of
A
the
is that the steel was consistently cleaner than any of the
other irons, and that the inclusions that were present were small
spheroidal inclusions.
These aspects are currently the
subject
of more detailed analysis.
The
shown
metallurgical
that
analysis of the Coppergate ironwork
the Anglo-Scandinavian smiths were
science of blacksmithing.
the
quality
of
experts
to
the some of the welds,
suggest
e.g.
the
that
this
level
of
presence
There is
expertise
consistent throughout the Anglo-Scandinavian Period.
knives
was
of
now
not
The Type 1
were usually manufactured from a steel core with piled or
phosphoric
iron
sharpening
has
migrated
cutting
also
the
Their expertise can be demonstrated by
white/yellow weld lines due to nickel enrichment.
evidence
in
has
sheaths,
but the evidence of wear
shown that in some cases the
away from the steel,
to the extent that the
edge was formed by one of the sheaths.
observed
in
the
cutting
spoon-bit.
The
and/or
re-
edge
had
effective
This effect was
correlation
with
archaeological typology (Table 2) showed the Group A knives
9
the
were
mostly of Type 2 manufacture and were of better quality. This may
have
been the result of the function of the Group A knives,
recent analysis of knives from Anglo-Saxon Southampton
earlier
date
than
but
(i.e.
Coppergate York) has shown that the
of
Type
2
manufacture is predominant, that the overall quality is very good
and
that there was very little use of phosphoric
therefore
suggested
that
the
Type
2 method
artefacts
from
iron.
of
It
is
manufacture
predates Type 1.
Conclusion
The
analysis
Coppergate,
were:
ore
is
of
the
Anglo-Scandinavian
York has shown that the four types of iron
employed
phosphoric iron, ferritic iron, piled iron and steel.
The
sources for the different types cannot be determined but
it
suggested that low phosphorus ores may not have been required
to manufacture the steel or the ferritic iron.
was
the
usages.
most widely used iron,
The
the
steel
but all the irons had
blacksmiths were highly skilled;
were readily achieved,
York
iron
specific
excellent
welds
and high hardnesses obtained for some
Anglo-
and Anglo-Saxon Southampton that the
earlier
edges.
There
is
evidence
of
from
cutting
Scandinavian
Phosphoric
material was manufactured to a higher standard.
It
term
has probably been noted that throughout this
'wrought
iron'
has not been used.
This is
paper
the
because
the
evidence given here, in particular, the manufacture of four types
of iron, makes the use of a single term inappropriate, since they
are all 'wrought'.
Note:
This work forms a part of a programme funded by the
Historic Buildings and Monuments Commission for England.
10
Fig 1
THE IRONWORKING PROCESS
'ORE TO ARTEFACT'
(THE SOURCES OF PROCESS DATA
ARE UNDERLINED)
!•Iter C. S1unders. The Iron Firedo9 florn We!wyn, Henford1hin,
I
I
100%
t
~ Smithing Residues
:.;;
~>
""~
~
vu"
£
~
.
Hen1. Ar<:hnofo•n II 1917 ll·<HJ
% OF ARTEFACTS
90%
I
10%
0.1%
0%
ARTEFACT~&so
/~
/
.
I
SMITHING
//BURIAL
~
~
c,O'<'
Ore Preparation
t
Mining/Collection
dRE
'
ARCHAEOLOGICAL ARTEFACT
/
\
~
/
~
. .....-- SME~ING
Smeltrng Resrdues
Rtcon~ideud,
;
I
\
/
I
~·
~
-ti;
os\0~
..... co~~''i~ p.·110
~·
o_,
- (,ol'lS
NATURAL WORLD--
Artefact Typology
"
Artefact Analys.is
Fig 2
Knife
Terminology
blade
tang
s2
knife back ~
cutting edg!! z
corrosion
s1
cutting edge section
s2
knife back
{?
t/b
"
tang/blade interface
w
weld line
v
4/b
J
;;w
Fig 3
D
0
Knife
Manufacturing
Ferrihc/ Phosphoric Iron
1
2
Typology
~Steel
3
4