Summary of Findings
There is no ISO or ASTM product standard for Söderberg electrode paste. YB/T 5215 is a product standard — it classifies grades and sets limit values — while the ISO and ASTM documents that buyers cite against it are test-method standards, and their stated scope is aluminium-industry carbon: green and calcined coke, pitch for electrodes, baked anodes and cathode blocks. The one significant product-standard counterpart is regional: ГОСТ Р 70859-2023, covering electrode paste for self-baking electrodes in electrothermal furnaces.
That structural asymmetry, not laboratory error, is the usual reason a single batch returns different numbers in a Chinese mill certificate and in a receiving laboratory abroad. The sections below map each parameter to its method standard on both sides, identify the four steps at which values actually diverge, and set out what a purchase contract must state so that the two results become comparable.
Key Findings
- No international product standard covers electrode paste. ISO and ASTM supply test methods only; ГОСТ Р 70859-2023 is the notable regional exception, applicable to paste for self-baking electrodes.
- The ISO and ASTM methods were written for a different material. ISO 8005:2005 addresses green and calcined coke, ISO 8006:1985 pitch for electrodes, and ISO 11713, ISO 18515 and ISO 12985 baked anodes and cathode blocks. Applying them to electrode paste is an extension beyond their stated scope, and the extension has to be defined in the contract.
- Two paste parameters have no international counterpart at all: volatile matter as measured by YB/T 5189—2022, and elongation as measured by YB/T 5289—2017. Both are written for the paste itself.
- Specimen preparation is the largest single source of divergence. Compressive strength and electrical resistivity are measured on a baked specimen; where the baking schedule is not agreed, the two laboratories are not testing the same material.
- The remedy is contractual, not analytical. Method number with edition year, specimen preparation, sampling plan, acceptance rule and arbitration laboratory must all appear in the purchase specification; a contract that says only “conforms to YB/T 5215” leaves every one of these open.
Contents
- Is there an international equivalent to YB/T 5215?
- Which method standard corresponds to each parameter?
- Where do divergent values actually come from?
- How should the test basis be written into a contract?
- What should be done when two laboratories disagree?
- Frequently asked questions
- Related reading
1. Is There an International Equivalent to YB/T 5215?
For Söderberg electrode paste, no. ISO and ASTM publish no product standard that classifies paste grades and sets limit values. What they publish are test-method standards, and the scope statements of those documents name aluminium-industry carbon materials — coke, pitch, baked anodes, cathode blocks — rather than self-baking electrode paste.
One regional counterpart exists. ГОСТ Р 70859-2023, Electrode paste for self-baking electrodes of electrothermal furnaces — Technical requirements, is a product standard in the same sense as YB/T 5215, and it governs in the CIS market. Buyers in Russia and Kazakhstan will normally specify against it; buyers in South Asia, Southeast Asia, the Middle East and South America generally specify against their own purchase documents.
The consequence is structural. Where the product-standard layer is empty, its function does not disappear — it transfers to the contract layer, and whichever party writes the specification defines what conformity means.
2. Which Method Standard Corresponds to Each Parameter?
The mapping below pairs each parameter with the method standards applied on either side, and states the scope of the international method. The scope column is the operative one: it is what determines whether a result is directly comparable or requires an agreed extension.
| Parameter | Chinese method standard | ISO | ASTM | Stated scope of the ISO/ASTM method |
|---|---|---|---|---|
| Ash | GB/T 1429—2009 (carbon materials) | ISO 8005:2005 (green and calcined coke); ISO 8006:1985 (pitch for electrodes) | ASTM C561 (ash in a graphite sample) | Coke, pitch or graphite — not a green paste |
| Volatile matter | YB/T 5189—2022 (electrode paste) | — | — | No international counterpart; the Chinese method is written for the paste itself |
| Compressive strength | GB/T 1431—2019 (carbon materials) | ISO 18515:2014 (cathode blocks and baked anodes) | ASTM C695 (carbon and graphite) | Requires a baked, machined specimen |
| Electrical resistivity | GB/T 24521—2018 (carbon raw materials and coke); GB/T 24525—2009 (graphite products) | ISO 11713 (cathode blocks and baked anodes; 2000 edition, revised 2025) | ASTM C611 (manufactured carbon and graphite, room temperature) | Machined specimen at ambient temperature |
| Bulk / apparent density | GB/T 24528—2009 (carbon materials) | ISO 12985-1:2018 (dimensions method); ISO 12985-2:2018 (hydrostatic method) | ASTM C559 (bulk density by physical measurements) | Two ISO parts give different values on porous material |
| Softening point | Per contract (binder pitch property) | — | ASTM D3104 (Mettler softening point of pitches) | Mettler and ring-and-ball methods return different numbers for the same pitch |
| Elongation | YB/T 5289—2017 (electrode paste) | — | — | No international counterpart; a paste-specific index |
Table 1. Only ash, compressive strength, resistivity and density have international method counterparts at all — and each counterpart was written for a different material, which is why the scope column governs the comparison.
Sulfur, fixed carbon and moisture are not tabulated here because several method families apply to each, drawn from coal and coke testing practice, and they return different values. Where these appear in a specification, the method must be named explicitly rather than left to the testing laboratory.
3. Where Do Divergent Values Actually Come From?
Four steps separate a batch from a reported number. Divergence accumulates across all four, but not evenly.
Sampling and division. Electrode paste is heterogeneous by construction — a binder phase distributed through granular aggregate. The sampling point, the number of increments and the division method affect the result before any instrument is involved. This step is treated in detail in How Sampling and Sample Preparation Affect Electrode Paste Test Results.
Specimen preparation. This is the largest single contributor. Compressive strength and electrical resistivity cannot be measured on green paste; a specimen must first be baked. Neither the Chinese nor the international method standards prescribe a baking schedule specific to electrode paste, because their scope assumes an already-baked article. Where heating rate, final temperature and hold time are not agreed between the parties, the two laboratories are measuring two different materials and no reconciliation of the numbers is possible.
Test conditions. Ashing end-temperature and ramp affect ash; specimen geometry and probe spacing affect resistivity; and for density, the dimensions method of ISO 12985-1 and the hydrostatic method of ISO 12985-2 return different values on a porous specimen, the difference widening with open porosity.
Acceptance rule. Whether conformity is judged on a single value or on the mean of several, how results are rounded, and whether a re-test is permitted after a failing result — none of this is fixed by a test-method standard. It belongs to the contract, and it is the step most often left unwritten.
4. How Should the Test Basis Be Written Into a Contract?
A specification that states only “conforms to YB/T 5215” leaves the entire method layer open. Six items close it:
- Method number with edition year for every parameter —
GB/T 1429—2009,ISO 11713:2025, and so on. An undated reference silently follows the standard through revision. - Specimen preparation, stated explicitly where the parameter requires a baked specimen: heating rate, final temperature, hold time, and specimen dimensions.
- Sampling plan: number of increments, sampling points across the consignment, division method, and retained-sample quantity.
- Acceptance rule: single value or mean, number of determinations, rounding convention, and the treatment of a borderline result.
- Arbitration laboratory, named in advance together with the party bearing the cost. Naming it after a dispute has begun rarely succeeds.
- Retention period for retained samples, at both ends, long enough to cover ocean freight plus the buyer’s inspection cycle.
Items 2 and 5 are the ones most often omitted, and they are the two that decide the outcome of a dispute.
A supplier specification may of course be tighter than the standard it references. Jiyang Carbon specifies standard-grade paste at 2.0%–5.0% ash, with premium grade to 1.8%, compressive strength above 18 MPa and electrical resistivity at or below 65 μΩ·m, and reports seven parameters per batch on a traceable Certificate of Analysis (COA). A tighter supplier limit does not remove the need for items 1 to 6 — it makes them more important, because a tighter limit is only meaningful once the method behind it is fixed.
5. What Should Be Done When Two Laboratories Disagree?
Work through the four steps of Section 3 in order, and stop at the first one that differs. Most disputes resolve before the fourth.
- Compare the method numbers and edition years actually used, not the ones named in the contract. A laboratory following its accreditation scope may have applied a different method from the one specified.
- Compare specimen preparation, in writing, from both laboratories. Where the baking schedules differ, the investigation ends here: the results are not comparable and a joint re-test on an agreed schedule is the only path forward.
- Compare sampling records against the agreed plan. Non-conformant sampling invalidates the result irrespective of how the test itself was run.
- Where the first three align and a material difference remains, submit the retained samples to the arbitration laboratory named in the contract, with both parties’ method statements attached.
Retained samples are what make this sequence possible. Where neither party has one, no step beyond the first can be completed, and the discussion becomes commercial rather than technical.
6. Frequently Asked Questions
Q1. The supplier’s report cites Chinese standards and our laboratory tests to ASTM. Which result governs? Neither, until the contract says so. The two results are not competing measurements of the same quantity — different method scopes and, usually, different specimen preparation. The contract should name one method set as governing and the other as informational. Where the contract is silent, the practical route is a joint re-test on an agreed method and an agreed specimen preparation, using retained material from the same batch.
Q2. Is “conforms to YB/T 5215” sufficient in a purchase order? It fixes the classification and the limit values, and nothing else. Sampling, specimen preparation, acceptance rule and arbitration remain undefined, and those are the four things a dispute turns on. The six items in Section 4 should be added, and they occupy one page.
Q3. Our specification cites ISO methods written for baked anodes. Is that defensible for electrode paste? It is workable, provided the extension is explicit. Add a clause stating that the method is applied to a specimen prepared from the paste under a defined baking schedule, and state that schedule. Without it, the specification cites a method whose scope does not cover the material being purchased, and either party can contest a result on that basis.
Q4. Should the purchase specification copy the standard’s limit values, or set its own? Set its own, at or inside the standard, driven by furnace type and operating practice. The standard is a floor for market-wide acceptability, not a specification for a particular furnace. Supplier limits are frequently tighter than the standard; the point of setting your own is that the limits then correspond to your operating window rather than to a national average.
Q5. A referenced standard has been revised. What changes? Undated references follow the revision automatically, which may move a limit or a method without any action by either party. Dated references do not, but then the contract keeps citing a superseded document. Either approach works if it is deliberate; the recommended practice is dated references plus a scheduled review of the specification, and a review interval of a few months is sufficient for standards of this kind.
7. Related Reading
- How Sampling and Sample Preparation Affect Electrode Paste Test Results — the detailed treatment of the first two divergence steps in Section 3
- From Incoming Materials to COA — how the batch record behind a test report is produced
- How to Review a Graphite Electrode Quality Package — the parallel review procedure on the graphite electrode side
- How to Compare Electrode Paste Supplier Offers Without Relying on One Test Value — why a single reported value cannot rank offers
- Söderberg Electrode Hard Breakage vs. Soft Breakage: A Root-Cause Diagnostic Guide — where a COA and a method dispute enter a field investigation
Author and Review
Written by Sterling Yang, Senior Vice President at Jiyang Carbon.
This article was reviewed by Mr. Yang, Technical Director at Jiyang Carbon, drawing on 40+ years of hands-on experience in carbon processing and on weekly review of paste production and test data.
Every standard number, title and scope statement in this article is taken from the issuing body’s published catalogue entry. Edition years are those current at the date of review and should be re-verified at each scheduled review, since several of these documents are under periodic revision. Limit values from YB/T 5215 are deliberately not reproduced here; a purchase specification should cite the standard text itself rather than a secondary source.
Technical Consultation
For assistance in drafting the method and acceptance clauses of an electrode paste purchase specification, or in reconciling two laboratories’ results on the same batch, please contact the technical team.
