Table 1 Benchmark alternatives to chromate with given prediction results examples.

From: Searching for chromate replacements using natural language processing and machine learning algorithms

#

Benchmark alternatives to chromate10,38,39

Examples from the NLP predictions

1

Trivalent chromium

2

Rare-earth-based inhibitors (aka lanthanide systems)

Cerium, CeN3O9

3

Vanadate based inhibitors

BiO4V (BiVO4), vanadate

4

Li-containing conversion coatings/primers

Lithium, LDH

5

Organic systems

Polyurethane, amines, BTA

6

Phosphate-based systems

Zinc phosphate, phosphates

7

Mg-rich primers

8

Zirconium conversion coatings

9

Titanium-containing conversion coatings

10

Silicon based systems

Silanes and sol–gel

11

Zinc-based coatings

12

Molybdate compounds

Molybdate, MoNa2O4 (Na2MoO4)

13

Calcium based systems

14

Electro-coatings/Electrophoretic systems

Electroplating, Co–P

15

Nanocomposites incl. nanoparticles

Alumina, ceria, titania, graphene

16

TFSAA (thin film sulfuric acid anodisation)

Anodising

17

BSAA (boric sulfuric acid anodising)

Anodising, sulfuric

18

Filled electroless nickel

19

WC/C (tungsten carbide carbon) coating

WC–Co, WC, (WCr)2C-Ni

20

Nitride coatings

TiN, CrxN, TiN/CrN

  1. The list of benchmark categories was derived from three research sources10,38,39. The prediction results in both Word2Vec and BERT models were manually identified, and part of the relevant materials was allocated here.