The invention concerns the field of manufacturing zirconium-alloy elements for 5 pressurised-water nuclear reactors, in particular structural tubes and cladding tubes for fuel for rods in nuclear fuel assemblies.
Various zirconium alloys - tertiary or quarternary (i.e. having two or three principal alloy elements in addition to Zr, respectively) - the composition of which may be coupled to a particular thermomechanical treatment and/or finishing method allowing the product they form
10 to be given increased corrosion resistance characteristics, are offered to users in order to produce pressurised-water nuclear reactor components. These alloys are used, inter alia, for structural components (gratings, guide tubes, and, where applicable, instrumentation tubes) and cladding tubes for fuel pellets, also known as sheaths, in nuclear fuel assemblies. These alloys must withstand various forms of corrosion that may occur during the normal operation
15 of the reactor, as well as good resistance to corrosion in accident conditions, in particular in the event of coolant loss (Loss Of Coolant Accident, LOCA), i.e. at very high temperatures (above 900°C) and in a water-vapour atmosphere.
It is known that high surface roughness on tubes for nuclear fuel assemblies degrades their corrosion resistance in the reactor.
20 Document WO-A-2006/027436, for example, has shown that a final mechanical
polishing step of the outer surface of a cladding tube, which gives it a roughness Ra less than or equal to 0.5 μm, coupled with a zircon alloy composition containing, in addition to zirconium and impurities resulting from production, 0.8 - 2.8% Nb, 0.015 -0.40% Fe, 600 -2300 O, 5 - 100, and, optionally, a small amount of Sn, Cr, or V, and a method for producing
25 the tube, allowed for an improvement in the tube’s corrosion resistance at high temperatures, in particular at the temperatures that may be found in the event of a LOCA. The need to limit the Hf and F content of the alloy to the extent possible is also affirmed, and the final mechanical polishing step allows for any trace F, which might result from pickling in a fluorinated bath, to be removed from the surface, whilst obtaining the desired roughness Ra.
30 The behaviour of a zirconium-alloy tube in the event of a LOCA is evaluated, for
example, by means of oxidation testing a sample of the tube in a water-vapour environment at a temperature of 1000°C. Such a test is described, e.g., in the paper ‘AREVA NP M5® Cladding Benefits for Proposed U.S. NRC RIA and LOCA Requirements’ presented in
2

September 2016 at the LWR Fuels with Enhanced Safety and Performance Meeting (TopFuel 2016).
The corrosion kinetics, measured by the increase in the mass of the sample resulting
from oxidation, is initially parabolic in nature. A deterioration of the kinetics (commonly known
5 in the field as ‘breakaway’) will occur after a certain test duration due to accelerated corrosion
and/or significant hydrogen absorption (‘hydride cracking’) (typically hydrogen absorption in
excess of 200 ppm).
The hydride cracking of a zirconium-alloy component degrades its mechanical and
microstructural properties, and may result in deformation or breakage, in whole or in part, e.g.
10 due to cracking, followed by local bursting in the case of a cladding tube for a nuclear fuel rod.
In absolute terms, tubes according to the recommendations of document WO-A-
2006/027436 have good corrosion resistance in accident conditions, with breakaway
occurring after approximately 5000 s, compared to approximately 1800 s in the case of more
commonly used alloys.
15 However, if an even greater delay of the occurrence of breakaway could be obtained,
this would represent a fundamental advantage for nuclear reactor safety in the event of accidents.
The objective of the invention is to propose a method that will allow for tubes for
nuclear fuel assemblies for pressurised-water reactors with improved corrosion and hydride
20 cracking resistance compared to known-art alloys, in particular M5 alloys, in particular in the
event of exposure to very high temperatures in accident conditions such as LOCA, to be
reliably obtained.
To this end, the invention concerns a tubular component for a pressurised-water
nuclear reactor, the composition by weight of which consists of:
25 - 0.8% ≤ Nb ≤ 2.8% ;
- traces ≤ Sn ≤ 0.65% ;
- 0.015% ≤ Fe ≤ 0.40% ; preferably 0.020 % ≤ Fe ≤ 0.35 % ;
- traces ≤ C ≤ 100 ppm ;
- 600 ppm ≤ O ≤ 2300 ppm ; preferably 900 ppm ≤ O ≤ 1800 ppm ; 30 - 5 ppm ≤ S ≤ 100 ppm ; preferably 8 ppm ≤ S ≤ 35 ppm ;
- traces ≤ Cr + V + Mo + Cu ≤ 0.35% ;
- traces ≤ Hf ≤ 100 ppm ;
- F ≤ 1 ppm ;
3

the remainder being zirconium and impurities resulting from production, the outer
surface of which has a roughness Ra less than or equal to 0.5 µm, obtained following a final
mechanical polishing step, characterised in that its outer surface has a roughness Rsk ≤ 1 in
absolute value and a roughness Rku ≤ 10.
5 The outer surface of the component may have a roughness Ra less than or equal to
0.3 µm, resulting from the final mechanical polishing step.
The outer surface of the component may have a roughness Rsk ≤ 0.75 in absolute value and a roughness Rku ≤ 9.
The invention also concerns a method for producing a fuel cladding tube for a nuclear 10 reactor, characterised in that:
- a zirconium-alloy ingot having the following composition by weight is prepared:
* 0.8% ≤ Nb ≤ 2.8% ;
* traces ≤ Sn ≤ 0.65% ;
* 0.015% ≤ Fe ≤ 0.40% ; preferably 0.020 % ≤ Fe ≤ 0.35 % ; 15 * traces ≤ C ≤ 100 ppm ;
* 600 ppm ≤ O ≤ 2300 ppm ; preferably 900 ppm ≤ O ≤ 1800 ppm ;
* 5 ppm ≤ S ≤ 100 ppm ; preferably 8 ppm ≤ S ≤ 35 ppm ;
* traces ≤ Cr + V + Mo + Cu ≤ 0.35% ;
* traces ≤ Hf ≤ 100 ppm ; 20 * F ≤ 1 ppm ;
the remainder being zirconium and impurities resulting from production;
- the ingot is subjected to forging, optionally followed by quenching, then extrusion
and thermomechanical treatments including cold rolling separated by intermediate annealings,
wherein all intermediate annealings are carried out at a temperature below the transus
25 temperature a → a + p of the alloy, ending with relief, semi-recristallisation or recristallisation annealing, and resulting in the production of a tube;
- optionally, chemical pickling and/or electrolytic polishing and/or initial mechanical
polishing of the outer surface of the tube are carried out;
- and final mechanical polishing of the outer surface to give it a roughness Ra less
30 than or equal to 0.5 µm, a roughness Rsk ≤ 1 in absolute value, and a roughness Rku ≤ 10 is
carried out.
Intermediate annealing may be carried out at temperatures no greater than 600°C. The final mechanical polishing step may be carried out with a finishing roller.
4

The final mechanical polishing step may be carried out by means of abrasion with an abrasive paste.
The final mechanical polishing may be carried out by a method selected from: honing,
abrasive paste extrusion, abrasion using a polishing felt or sheet impregnated with abrasive
5 paste.
The final mechanical polishing step may be carried out by roller burnishing.
As has been made clear, the invention consists of producing a tubular component for
a pressurised-water nuclear reactor, in particular a structural tube, i.e. a guide or
instrumentation tube, from a tube made of a Zr-Nb alloy with 0.8 – 2.8% Nb, also containing a
10 small amount of Fe and S, as well as Sn, Cr, V, Mo, and/or Cu, and having an O content that
may be relatively high, and prepared by the method described in WO-A-2006/027436, with
the possible exception of the post-forging quenching, which is not strictly necessary with Zr-
Nb alloys. After having subjected it to thermal treatments adequate to give it the desired
mechanical properties, and, preferably, a chemical pickling step (generally carried out before
15 the final thermal treatment), the outer surface of the tube is polished by a method allowing for
a particular surface finish, defined not only by its Ra value, but also by the Rsk and Rku
values to be obtained as a result of a mechanical polishing operation (‘final mechanical
polishing’). These requirements serve to ensure that the outer surface of the tube will have a
morphology that will render it as insensitive as possible to corrosion and/or hydride cracking
20 in accident situations, in particular in the event of LOCA.
Other polishing operations, which may not necessarily all be mechanical, may
precede the final mechanical polishing step that results in the type of roughness according to
the invention, thus constituting the main step of the method according to the invention. In the
following, ‘initial polishing’ will refer to a polishing step that merely constitutes an intermediate
25 step on the way to obtaining the desired roughness, and ‘final polishing’ will refer to the last
polishing step, which results in the desired roughness.
It goes without saying that, if a single mechanical polishing operation is carried out
during the treatment according to the invention of the surface of the product, this operation
will constitute the ‘final mechanical polishing’ step. This final mechanical polishing step may
30 be followed by other production steps, e.g. inspection, degreasing, etc., but none of the other
steps must result in surface contamination, in particular with halogens, or a degradation of its roughness.
5

Based on its composition, in particular the alloy produced by Framatome, known by the brand names ‘M5’ or ‘M5Framatome’ falls within the scope of the invention.
The invention will be better understood based on the following direction, with
reference to the following appended drawings:
5 Fig. 1, showing the increase in mass of reference samples of a Zr-Nb alloy
(M5Framatome) consistent with the composition and Ra requirements of WO-A-2006/027436 as a function of the square root of the time spent at a temperature of 1000°C in a water vapour environment;
Fig. 2 shows the development of the hydrogen content of the same reference samples
10 as a function of the square root of the time spent at a temperature of 1000 °C in a water
vapour environment.
Fig. 3 shows the increase in mass and hydrogen content of the same reference
samples and of samples according to the invention as a function of the square root of the
time spent at a temperature of 1000 °C in a water vapour environment.
15 The behaviour of a tube during a LOCA test as described above depends on the
roughness of the outer surface, which is most commonly described solely based on the Ra
parameter as defined in standard NF EN ISO 4287. This corresponds, over a given
evaluation length (‘base length’) to the arithmetical mean deviation of the roughness profile of
the surface, which includes protrusions and cavities of varying heights relative to the mean
20 line of the roughness profile. Ra constitutes an evaluation of the mean of the absolute values
of the heights. Ra is calculated by the following formula:

where lr is the base length of the roughness profile and Z(x) is the ordinate (or height)
of the roughness profile for an abscissa x on the mean line of the roughness profile. It should
25 be noted that the origin of the height is the mean value of the roughness profile, and that,
accordingly, the integral of Z(x) taken from 0 to lr is nil.
In fact, however, the experience of the inventors has shown that the parameter Ra is
insufficient in order to fine-tune the behaviour of the alloy in conditions likely to give rise to
significant oxidation and/or hydride cracking of the tube, and, in particular, to explain the very
30 good behaviour observed when the outer surface thereof has been treated according to the
invention.
6

5
10

The inventors have found that two parameters, defined in standard NF EN ISO 4287, were also of particular importance in solving the problem presented. These parameters are Rsk (‘skewness’) and Rku (‘kurtosis’).
The parameter Rsk defines the asymmetry of the roughness profile evaluated. It translates the asymmetrical height distribution relative to the mean line of the roughness profile, defined based on the base length lr. It provides information on the morphology of the surface state. A nil Rsk value corresponds to a normal (Gaussian) distribution of heights about the mean line. A positive Rsk value corresponds to a ‘hollow’ profile with a height distribution biased towards higher values, e.g. in the case of a plateau surface with a preponderance of protrusions. A negative Rsk value corresponds to a ‘full’ profile with a height distribution lower towards higher values, e.g. in the case of a plateau surface with a preponderance of cavities. Rsk is calculated by the following formula:

in which Rq is the mean quadratic deviation of the profile evaluated over the base
15 length lr according to:

20
25

Rq corresponds to the quadratic mean of the heights over the base length lr.
The parameter Rku defines the kurtosis of the roughness profile under evaluation, i.e. the breadth of the height distribution relative to the mean line of the roughness profile, defined based on the base length lr. It provides information on the morphology of the surface state. An Rsk value equal to 3 corresponds to a normal (Gaussian) height distribution. An Rku value greater than 3 corresponds to a ‘dense’ profile relative to the normal distribution, i.e. predominantly having heights with low absolute value relative to the mean line of the roughness profile. An Rku value less than 3 corresponds to a ‘staggered’ profile relative to the normal distribution, i.e. with a greater proportion of heights far from the mean line, e.g. with heights equally distributed over the entire span. Rku is calculated by the following formula:

7


In particular, Rsk and Rku are used in tribology to evaluate the contact, wear
resistance, and lubrication properties of the surface measured, but they are not used to
evaluate the corrosion resistance of a surface.
5 The inventors have found that, all things being equal, if the parameters Rsk
(skewness) and Rku (kurtosis) of the outer surface of the tube meet certain criteria, the oxidation kinetics in accident conditions, in particular in the event of LOCA, remains parabolic over the duration of the test. Otherwise, oxidation accelerates over the course of the test.
The macroscopic surface constraints caused by oxidation alone cannot explain the 10 difference in behaviour observed on samples of tubes with delayed - or no - breakaway over the duration of the test. The hypothesis advanced by the inventors to explain this difference in behaviour is that the constraints could also have a local effect on oxidation on the level of surface roughness. A surface with numerous pronounced protrusions may have an increased risk of cracking of the oxide perpendicular to the oxide-metal interface, and locally 15 accelerated oxidation at the location of the protrusions.
The desired surface is a polished surface (Ra ≤ 0.5 µm, preferably ≤ 0.3 µm) having a
substantially symmetrical roughness distribution, i.e. a skewness factor Rsk near nil in
absolute value: |Rsk| ≤ 1 and preferably |Rsk| ≤ 0.75, and no pronounced protrusions or
cavities, which translates into a kurtosis factor Rku less than or equal to 10, preferably less
20 than or equal to 9.
The improved behaviour observed may be obtained reproducibly by carrying out careful mechanical surface finishing, which allows the desired roughness to be obtained in the outer surface of the tube.
Because this finish can be obtained by various means, they will not be described 25 exhaustively herein.
One possible manner of obtaining this finish consists of successively polishing the
tube with silicon carbide SiC rollers of increasing grain sizes (e.g. up to grain 240 mesh or
more according to ISO 8486-2), with these operations constituting an initial mechanical
polishing step, and ending with a final polishing step using a finishing roller such as a rolled
30 finishing wheel, a radial brush, a flap disk with a very fine grain, e.g. a Scotch Brite™ finishing
8

roller. This method of finishing allows for tubes to be obtained that, at a minimum, have
delayed breakaway, i.e. occurring after more than 10,000 seconds, for an alloy having a
composition and method of preparation prior to the final polishing step that are substantially
consistent with those set forth in WO-A-2006/027436.
5 The initial polishing step may also comprise non-mechanical polishing (e.g. chemical
or electrolytic polishing), used alone or in combination with mechanical polishing. This initial polishing step is then followed by a final mechanical polishing operation.
If experience shows that a single mechanical polishing step allows the desired roughness to be obtained in the product in question, it is possible to carry out only a single
10 mechanical polishing step, which will be termed a ‘final mechanical polishing’ step because it
does constitute the last polishing operation carried out on the surface of the product.
The mechanical polishing steps and the means used for these steps, in particular for the final mechanical polishing step, may be determined with providers of this type of equipment based on a specification, which conventionally includes the desired final
15 roughness and the method for evaluating it. This will also specify the polishing materials that
should be avoided as potentially harmful or difficult to remove, in particular those listed in applicable documents such as the RCC-C (Design and Construction Rules for Fuel Assemblies of PWR Nuclear Power Plants) published by AFCEN (Association Française pour les Règles de Conception, de Construction et de Surveillance en Exploitation des Matériels
20 des Chaudières Electro-Nucléaires).
The exact parameters of the mechanical polishing processes, initial and final where applicable, that allow for the type of roughness desired may be determined experimentally by persons skilled in the art by means of a standard test series. To this end, it will be necessary to correlate the polishing means to be used, as well as their use parameters, with the
25 composition of the tube and the thermomechanical treatments to which it has been subjected,
as well as any chemical pickling and/or electrolytic polishing that may have preceded the mechanical polishing step(s). In particular, all things being equal, these characteristics affect the hardness and the state of the outer surface of the tube before the mechanical polishing step(s), and may also play a role in the result of the final mechanical polishing step.
30 This method for surface finishing the tube to obtain the fuel sheath according to the
invention is thus applied to a zirconium-alloy tube, which may contain impurities resulting from production, the composition by weight and method of preparation are as follows for the reasons stated in WO-A-2006/027436.
9

Its Nb content is 0.8 – 2.8 %.
Its Sn content ranges from trace (in other words, a content equal to nil or just barely
above nil, resulting merely from the production of the alloy without any intentional addition of
the element in question) to 0.65 %. The normal detection limit of this element is
5 approximately 30 ppm, and it should be understood that the Sn content may go down to
values corresponding to mere traces as defined above (thus including a value that would be strictly nil).
Its Fe content is at least 0.015%, preferably at least 0.02%, and no more than 0.40%,
preferably no more than 0.35%.
10 Cr, V, Cu, or Mo may be optionally present in order to supplement or supplant part of
the Fe, provided that the sum of their content does not exceed 0.35 %.
The C content of the alloy may not exceed 100 ppm.
The alloy contains between 600 and 2300 ppm O, preferably between 900 and 1800
ppm.
15 The S content must be kept between 5 and 100 ppm, preferably between 8 and 35
ppm.
The presence of Hf in the alloy should be avoided. Hf content should be very low,
such that, in the final alloy, no more than 100 ppm Hf, preferably no more than 75 ppm Hf is
present. Particular attention should be paid to separating Hf during the preparation of the Zr
20 sponge from which the alloy is produced.
Any F in the alloy should be limited to no more than 1 ppm.
Another very important requirement is the absence of fluorides on the alloy surface.
As noted in WO-A-2006/027436, in order to obtain a structural or cladding tube with
improved corrosion and hydride cracking resistance in the case of a LOCA, it is absolutely
25 indispensable to use surface preparations that result in radical elimination of fluorides. From
this point of view, the mechanical polishing subsequent to chemical pickling is the most
suitable method for preparing the tube surface before use.
Furthermore, there is the risk that the highly precise requirements for the roughness
characteristics of the outer surface of the final tubular product, which are not limited to a
30 maximum Ra value, will not be easily met by means of chemical polishing. Thus, it is
necessary to carry out at least the final step of preparation of the tube surface in the form of polishing by mechanical means, e.g. by the method described above, examples of which will be provided below.
10

The preparation of tubes from the ingot resulting from the production of the alloy is carried out by a method including forging, optionally followed by quenching, spinning, and cold rolling steps separated by intermediate annealing steps, with all annealing being carried out at a temperature below the transus temperature a -► a + p of the alloy, thus generally 5 below 600 °C. These relatively low-temperature thermal treatments allow for good corrosion resistance under normal operating conditions, and include a final relief annealing, semi-recystallisation, or recrystallisation step, depending on the microstructure desired for the final product. This may differ for the various categories and different uses of the tubes falling within the scope of the invention. For example, recrystallisation is advisable if good stress 10 resistance is desired for the tube.
Generally, in industrial practice, it is advisable to carry out 3, 4, or 5 cold rolling passes, separated by intermediate annealing steps each carried out at a temperature between 500 and 580°C, e.g. between 1 h at 500°C and 12 h, or 24 h, at 580°C.
Another necessary condition for solving the problem presented is that the outer 15 surface of the tube is given very low roughness Ra, less than or equal to 0.5 µm, preferably less than 0.3 µm.
WO-A- 2006/027436 advises obtaining a roughness level Ra this low. However,
according to this invention, two other conditions are necessary in order to further optimise the
LOCA behaviour of the alloy in question:
20 - A Rsk value in absolute value less than or equal to 1 (thus between - 1 and
+1), preferably less than or equal to 0.75 in absolute value (thus between -0.75 and +0.75);
And an Rku value less than or equal to 10, preferably less than 9.
The invention seeks to obtain a significant extension of the period in which no
breakaway is observed for Zr-Nb alloy tubes including the M5Framatome alloy.
25 To this end, samples of cladding tubes (9.5 mm in diameter and 0.57 mm thick) were
tested for various compositions and outer surface roughness configurations, obtained, in particular, by means of final polishing steps of various types that will be discussed in detail below.
The Zr used for the production of the tubes was obtained by conventional methods in
30 the form of a sponge or low-Hf electrolytic crystals (less than 100 ppm in the alloy). Following
sufficient smelting to allow for the elimination of any residual fluorine (F < 1 ppm in the
finished tube), a conventional method for transforming the ingot to obtain cladding tubes,
guide tubes, or instrumentation tubes for pressurised-water nuclear reactors was used:
11

forging, optional quenching, pilgering in 3 - 5 passes with intermediate annealing steps at a
temperature below the transus temperature α → α + β. With the exception of quenching,
which was not systematically carried out, this method is identical to that described in WO-A-
2006/027436, in particular as regards the optional pickling and internal polishing steps.
5 Table 1 shows the compositions of 8 samples of these M5Framatome alloy tubes, the
manufacturing variants used, as well as their increases in mass and their hydrogen content, in connection with fig. 1 and 2 and/or fig. 3. All tubes are in the recrystallised state, and were pickled prior to the first thermal treatment.
Fig. 1 and 2 show the behaviour of reference samples of M5Framatome tubes consistent
10 with the composition requirements of WO-A-2006/027436, having the following composition, in a water vapour environment at 1000°C (oxidation test as described in the TopFuel 2016 cited supra): Zr ; 1,02% Nb ; between 200 and 1000 ppm Fe ; between 1000 and 1500 ppm O ; between 5 and 35 ppm S, and less than 1 ppm F, roughness Ra below 0.5, but not consistent with the invention in terms of the roughness Rsk, with values in certain cases in
15 the ranges [-1.65 ;-1] or [+1 ; +1.48], and/or roughness Rku with values in some cases in the range [10 ; 15.55].
Fig. 1 shows the increases in mass (due to oxidation) as a function of the square root of the residence time in the environment in question, and fig. 2 shows the development of the hydrogen content as a function of the square root of the residence time in the environment in
20 question (NB: Given that the square root of the residence time is reported on the abscissa axis, the curves are significantly flatter than they would be if the abscissa axis represented residence time).
Based on standard criteria, the reference samples have good corrosion and hydride cracking resistance in accident conditions, with breakaway occurring after approximately
25 5000 s, which results in rapid acceleration of oxidation (fig. 1) and hydride cracking (fig. 2), as shown by the position of the experimental points, which are consistently above the extensions (dotted line) of the regression lines representing the increase in mass (fig. 1) and H content (fig. 2) before breakaway occurs. Typically, as shown in fig. 1 and 2, the duration for which a fuel sheath is subjected to LOCA is 1800 s, but the sheath must be able to withstand longer
30 exposures.
In order to simplify the description, only 4 reference samples out of those tested are shown in table 1: samples 1, 4, 5, and 7.
12

For all samples in table 1, the nominal compositions are indicated as regards the main
alloy elements. They all contain 1.0 % Nb and an Fe content between 0.02 and 0.07%. All
tubes tested comprised less than 100 ppm C and Hf and less than 1 ppm fluorine. All
elements not mentioned are, at most, present in trace amounts.
5 The tubes of samples 1 – 8 all underwent 4 rolling passes with 2 h intermediate
annealing at 580 °C.
Table 1 also shows the results of measurements of Ra, Rku, and Rsk roughness
carried out using a Mitutoyo SV2000 roughometer on these cladding tubes. These roughness
values were obtained with various finishing means. The measurements were carried out in
10 accordance with the applicable standard. For example, for polishing marks running tangential
to the cladding tube, the measurements were carried out on tube generators over a length of 4 mm with a cut-off of 0.8 mm. Three measurements were carried out on each of the tubes; the mean and the standard deviation of these measurements are shown in table 1.
[Table 1]
Tube 1 2 3 4 5 6 7 8

Composition Zr
1.0Nb
0.02Fe Zr
1.0Nb
0.02Fe Zr
1.0Nb
0.05Fe Zr
1.0Nb
0.05Fe Zr
1.0Nb
0.07Fe Zr
1.0Nb
0.07Fe Zr
1.0Nb
0.04Fe Zr
1.0Nb
0.04Fe
O (wt%) 0.13 0.13 0.14 0.14 0.15 0.15 0.11 0.11
S (ppm) 8 8 13 13 32 32 22 22
Quenched No No Yes Yes No No No No
Initial polishing SiC roller SiC roller up to 240 SiC strip up to 240 SiC strip SiC blasting SiC
blasting up to 240 SiC roller SiC roller up to 240
Final polishing SiC
roller
240 Finishing roller Finishing roller SiC strip 240 SiC
blasting
240 Colloidal
siliocon
(sheet) SiC roller 120 Finishing roller
Mass
increase
(mg/cm²) 15.62 11.12 9.64 11.42 15.76 9.25 22.14 12.52
H content (ppm) 635 16 13 282 918 15 954 65
Residence time (s) 10 000 15 000 10 000 8 600 15 000 10 000 18 000 18 000
Ra (µm) 0.13 0.12 0.14 0.32 0.34 0.14 0.23 0.23
Standard deviation on Ra (µm) 0.01 0.01 0.02 0.02 0.01 0.01 0.00 0.01
Rsk -1.65 0.72 0.29 -1.21 0.75 -0.35 1.28 -0.53
Standard deviation on Rsk 0.06 0.88 0.48 0.79 0.64 0.97 0.89 0.78
Rku 6.32 8.71 8.96 6.08 11.55 6.91 10.32 4.69
13

Standard deviation on Rku 0.31 3.22 2.66 0.33 5.43 2.64 3.34 2.84
Consistent with invention No Yes Yes No No Yes No Yes
Table 1 : Composition, manufacturing variant, mass increase, hydrogene content, and
roughness of tubes 1 - 8
5 Tube 1 is a reference tube (the absolute value of its Rsk is too high), the roughness of
which was measured following polishing with silicon carbide rollers of increasing grain (initial mechanical polishing) up to a grain of 240 (final mechanical polishing). It has a roughness Ra substantially equal to that of tube 2 (itself consistent with the invention in all respects) from the same batch, which underwent the same polishing steps with silicon carbide rollers of
10 increasing grain up to a grain size of 240 (initial mechanical polishing), followed by final
mechanical polishing with the finishing roller.
Tube 3 (consistent with the invention), from a different lot to tubes 1 and 2, with a somewhat increased Fe content, underwent the same polishing steps as tube 2, except that the initial mechanical polishing was carried out with SiC strips of increasing grain (up to a
15 grain size of 240) in lieu of polishing with silicium carbide rollers of increasing grain. Tube 4,
from the same batch, underwent the same polishing steps with SiC strips of increasing grain (initial polishing) as tube 3 up to a grain size of 240 (final mechanical polishing). It did not undergo the final polishing step with the finishing roller, unlike tube 3, and it is not consistent with the invention because its Rsk value is somewhat too high.
20 Tube 5, from another batch with an even higher iron content, did not undergo the
initial mechanical polishing steps with rollers or strips, but with blasting with SiC grains of decreasing size. It underwent final mechanical polishing by blasting with SiC 240 grains. Tube 6, from the same batch, was additionally polished by rubbing with a polishing sheet impregnated with abrasive paste (colloidal silicon in this example) at the end. The Rku of tube
25 5 is too high, whilst tube 6 is consistent with the invention.
To confirm the impact of the initial polishing step, the initial polishing of tubes 7 and 8, from a batch with a medium iron content, with silicon carbide rollers of increasing grain was stopped at a grain size of 120. As expected, the duration of the final polishing step with the finishing roller on tube 8 had to be extended in order to obtain a roughness value consistent
30 with the invention, but it was possible. Roughness values consistent with the invention thus
14

do not depend entirely either on the initial polishing step, nor on the characteristics of the
instrument used for the final mechanical polishing step. Persons skilled in the art are able to
determine experimentally the conditions of the final mechanical polishing step (characteristics
of the polishing tool and parameters of its use, coupled with the duration of the polishing step)
5 that will allow the roughness according to the invention to be obtained.
Tube 1’s Rsk value is too high, although its Rku value is consistent with the invention and its Ra value is consistent with the invention and substantially equal to that of tube 2. Tube 7’s Rsk and Rku values are too high, although its Ra is consistent with the invention and equal to that of tube 8. This clearly shows that the three representative values for the
10 roughness of the tube are not strongly correlated, and that the final mechanical polishing step
has a quite particular importance in obtaining the precise roughness configuration according to the invention.
The behaviour of tubes 2, 3, 6, and 8 in table 1 in LOCA testing is shown in fig. 3. To facilitate the comparison, the results of the samples from fig. 1 and 2 were also shown
15 (shaded) in fig. 3.
For two samples from the same tube as sample 2, the trial was extended to 30,000 and 35,000 s respectively without any breakaway. The corresponding points are shown in fig. 3.
Fig. 3 also includes the results obtained for tubes 11 – 19 of table 2 below, the
20 compositions and roughness values of which are described. These tubes are distinguished
from those of table 1 by a greater alloy element contet, but their compositions remain consistent with the requirements of the invention. All tubes tested comprised less than 100 ppm C and Hf and less than 1 ppm fluorine. All elements not mentioned are, at most, present in trace amounts.
25
15

Tube 11 12 13 14 15 16 17 18 19
Composition Zr 1,0Nb 0,3Sn 0,1Fe Zr 2,0Nb 0,2Sn 0,2Fe Zr 0,8Nb 0,2Sn 0,12Fe Zr 1,0Nb 0,5Sn 0,1Fe 0,2Cu Zr 1,8Nb 0,6Sn 0,2Fe 0,1V 0,1Cr Zr 2,8Nb 0,4Sn 0,01Fe 0,3Mo Zr 1,5Nb 0,05Fe Zr 1,0Nb 0,3Sn 0,1Fe Zr 1,0Nb 0,3Sn 0,1Fe
O (ppm) 2248 1348 1048 694 1242 843 1129 1829 1099
S (ppm) 6 26 15 29 19 95 55 24 95
State Recrystallised Semi-recrystallised Recrystallise d Recrystallise d Recrystallise d Recrystallise d Relieved Relieved Semi-recrystallised
Pickling Yes Yes No Yes Yes Yes No Yes Yes
Initial polishing SiC roller 240 Chemical SiC roller 240 SiC strip 240 SiC strip 240 Chemical SiC roller 240 No Chemical
Final polishing Finishing roller Finishing roller Honing with
abrasive
paste Roller burnishing Finishing roller Abrasive paste (felt) Chemical Finishing roller SiC roller 240
Mass increase (mg/cm²) 10,67 11,35 11,28 9,67 10,06 12,27 13,72 12,03 16,35
H content (ppm) 16 16 61 18 45 15 383 82 474
Residence time (s) 10 000 15 000 18 000 10 000 10 000 15 000 10 000 18 000 15 000
Ra (µm) 0,16 0,12 0,12 0,14 0,15 0,12 0,08 0,11 0,38
Rsk 0,24 0,61 0,03 -0,25 0,24 -0,49 -0,02 -0,67 0,89
Rku 9,02 4,08 3,79 6,78 8,21 4,82 10,88 6,62 12,17
Consistent with invention Yes Yes Yes Yes Yes Yes No Yes No
Table 2 : Composition, manufacturing variant, mass increase, hydrogene content, and
roughness of tubes 11 - 19
16

Tubes 11, 13, and 17 all underwent the conventional range of initial mechanical
polishing with SiC rollers up to a grain size of 240; tubes 14 and 15 underwent initial
mechanical polishing with SiC strips up to a grain size of 240, and tubes 12, 16, and 19 initial
underwent chemical polishing, and tube 18 did not undergo any initial polishing step. The final
5 polishing step differs, as shown in table 2: chemical polishing or mechanical polishing by
various means: finishing roller, abrasion with abrasive paste (colloidal silicon, artificial diamonds, metal oxides of Ti or Zr), roller burnishing. For tube 19, the final mechanical polishing step was polishing with a 240-grain SiC roller, and its Rku is too high to be consistent with the invention. The final chemical polishing of tube 17, not followed by final
10 mechanical polishing, did not allow for an Rku value consistent with the invention. The
abrasive paste abrasion methods tested are honing with an abrasive paste containing synthetic diamonds for tube 13, and a felt impregnated with a mixture of metal oxides (Ti and Zr) for tube 16. Other abrasive paste abrasion methods could be used, e.g. milling by abrasive paste extrusion, or without using abrasive paste, as with tube 14 (roller burnishing).
15 Tube 8 directly underwent polishing with a finishing roller following the final thermal treatment.
As expected, the duration of the final mechanical polishing with the finishing roller had to be drastically extended in order to obtain the Ra, Rsk, and Rku characteristics required by the invention; thus, the method was poorly suited to industrial application. The examples of table 3 show that roughness values consistent with the invention do not depend on the presence or
20 nature (mechanical or non-mechanical) of initial polishing, and that the final mechanical
polishing step may be carried out by various means.
Fig. 3 shows that tubes produced according to the invention, in terms of composition and surface roughness, do not experience breakaway before a duration of exposure to 1000 °C water vapour that is in all cases significantly greater than the 5000 s known from the
25 prior art for similar alloys; see the grey points and the black points located above the
regression line in fig. 3, which correspond, respectively, to the reference samples of fig. 1 and 2 and samples 17 and 19 of table 2. In particular, the in the mass increase (corresponding to an acceleration of corrosion) and hydride cracking gradients (hydrogen recovery in excess of 200 ppm) are substantially delayed beyond 10,000 s. No difference in the very clear results
30 can be seen between the various ways of carrying out the final polishing step.
Although it was known that high-temperature oxidation resistance depended on the state of the surface, in particular the absence of fluorine pollution (pollution resulting, e.g., from pickling in a fluoronitric bath) and a controlled roughness Ra value, it was not known that
17

other roughness parameters related to the shape of the peaks and located below threshold values, i.e. Rsk and Rku, could allow for further delays in the kinetic acceleration of oxidation and hydride cracking of the tube in the LOCA context if chosen wisely.
The parameters Rsk and Rsku correspond to an analysis of the roughness
5 measurements carried out with 2D profilometry, i.e., an analysis of the geometric differences
of the state of the surface compared to the mean line. In the case of 3D profilometry
measurements, the equivalent parameters, Ssk and Sku, may be used, or the analysis may
be carried out on one or more generators rather than the surface as a whole.

WE CLAIMS

1. Tubular component for a pressurised-water nuclear reactor, having the following 5 composition by weight:
- 0.8% ≤ Nb ≤ 2.8% ;
- traces ≤ Sn ≤ 0.65% ;
- 0.015% ≤ Fe ≤ 0.40% ; preferably 0.020 % ≤ Fe ≤ 0.35 % ;
- traces ≤ C ≤ 100 ppm ;
10 - 600 ppm ≤ O ≤ 2300 ppm ; preferably 900 ppm ≤ O ≤ 1800 ppm ;
- 5 ppm ≤ S ≤ 100 ppm ; preferably 8 ppm ≤ S ≤ 35 ppm ;
- traces ≤ Cr + V + Mo + Cu ≤ 0.35% ;
- traces ≤ Hf ≤ 100 ppm ;
- F ≤ 1 ppm ;
15 the remainder being zirconium and impurities resulting from production, the outer
surface of which has a roughness Ra less than or equal to 0.5 µm, obtained following a final mechanical polishing step, characterised in that its outer surface has a roughness Rsk ≤ 1 in absolute value and a roughness Rku ≤ 10.
20 2. Tubular component for a pressurised-water nuclear reactor according to claim 1,
characterised in that its outer surface has a roughness Ra less than or equal to 0.3 µm, obtained following the final mechanical polishing step.
3. Tubular component for a pressurised-water nuclear reactor according to claim 1 or
25 2, characterised in that its outer surface has a roughness Rsk ≤ 0.75 in in absolute value and
a roughness Rku ≤ 9.
4. Method for producing a fuel cladding tube for a nuclear reactor, characterised in
that:
30 - a zirconium-alloy ingot having the following composition by weight is prepared:
* 0.8% ≤ Nb ≤ 2.8% ;
* traces ≤ Sn ≤ 0.65% ;
* 0.015% ≤ Fe ≤ 0.40% ; preferably 0.020 % ≤ Fe ≤ 0.35 % ;
19

* traces ≤ C ≤ 100 ppm ;
* 600 ppm ≤ O ≤ 2300 ppm ; preferably 900 ppm ≤ O ≤ 1800 ppm ;
* 5 ppm ≤ S ≤ 100 ppm ; preferably 8 ppm ≤ S ≤ 35 ppm ; 5 * traces ≤ Cr + V + Mo + Cu ≤ 0.35% ;
* traces ≤ Hf ≤ 100 ppm ;
* F ≤ 1 ppm ;
the remainder being zirconium and impurities resulting from production;
- the ingot is subjected to forging, optionally followed by quenching, then to extrusion
10 and thermomechanical treatments including cold rolling separated by intermediate annealings,
wherein all intermediate annealings are carried out at a temperature below the transus temperature a → a + p of the alloy, ending with a relief, semi-recristallisation or recristallisation annealing, and resulting in the production of a tube;
- optionally, chemical pickling and/or electrolytic polishing and/or initial mechanical
15 polishing of the outer surface of the tube is carried out;
- and final mechanical polishing of the outer surface to give it a roughness Ra less
than or equal to 0.5 µm, a roughness Rsk ≤ 1 in absolute value and a roughness Rku ≤ 10 is
carried out.
20 5. Method according to claim 4, characterised in that the intermediate annealings are
carried out at temperatures less than or equal to 600°C.
6. Method according to either of claims 4 or 5, characterised in that the final
mechanical polishing is carried out with a finishing roller.
25
7. Method according to either of claims 4 or 5, characterised in that the final
mechanical polishing is carried out by abrasion with an abrasive paste.
8. Method according to claim 7, characterised in that the final mechanical polishing is
30 carried out by a method selected from: honing, abrasive paste extrusion, abrasion using a
polishing felt or sheet impregnated with abrasive paste.
20

9. Method according to either of claims 4 or 5, characterised in that the final mechanical polishing is carried out by roller burnishing.