Low-Titanium Ferrophosphorus: Titanium Control Limits

Ferrophosphorus is one of the few ferroalloys whose specification is written around an impurity rather than around its main element. Buyers who need phosphorus in a steel or iron melt accept the 23-27% P without much debate; what they argue about is titanium, because titanium is the one element in the alloy that cannot be corrected downstream. This article explains where the titanium in ferrophosphorus comes from, what ceiling a producer can realistically guarantee, how the different levels behave in the melt shop, and how to write an acceptance clause that can actually be verified.

What Ferrophosphorus Is and Why Titanium Decides the Grade

Ferrophosphorus, often shortened to FeP, is an iron-phosphorus alloy containing roughly 23-27% phosphorus, 1-3% silicon, up to about 1% carbon and small amounts of manganese, sulphur and titanium. Metallurgically it is not a single compound. At commercial phosphorus levels the solidified alloy consists mainly of the phosphide Fe3P together with higher phosphides and a limited amount of iron-rich solid solution, which is why the material is extremely brittle, dull grey to dark grey on a fresh fracture, and almost impossible to machine. In the Fe-P binary system a eutectic forms near 1050 C at about 10.5% P and the Fe3P phase melts in the region of 1166 C, so a 24% P commercial alloy softens and melts over a range and is fully fluid well below 1450 C. That relatively low melting range is the reason the alloy dissolves quickly in a ladle or in the furnace heel, and the reason it can also fuse into lumps if a consignment is stored too warm or too damp.

The density of cast ferrophosphorus is high, in the region of 6.0-6.5 g/cm3, so the alloy sinks through slag and through liquid steel. This is a practical advantage during tapping and a practical problem during production, because the alloy collects at the bottom of the furnace and has to be tapped separately from the calcium-silicate slag. It is also the reason ferrophosphorus is recovered as a co-product of the electric phosphorus furnace rather than being melted deliberately: in the thermal process, phosphate rock, silica and coke are smelted at roughly 1400-1500 C to release phosphorus vapour, and the iron fed with the burden combines with part of the phosphorus to form a dense ferrophosphorus layer under the slag.

Its functions in the melt shop are narrow but not trivial. It is the standard carrier for deliberate phosphorus alloying, it improves fluidity in phosphoric cast irons, it acts as a deoxidiser and alloying addition in some copper-base alloys, and in powder metallurgy it is used as a sintering activator because phosphorus promotes interparticle bonding and increases the strength of sintered iron compacts. In none of these applications is titanium wanted.

Where the Titanium Comes From and Where It Goes in the Furnace

Titanium is not added to ferrophosphorus on purpose. It arrives with the burden. Phosphate rock carries a small amount of titanium as an oxide along with clay and iron minerals; the iron units, whether mill scale, low-grade ore or scrap, carry more, and titanium-bearing scrap is the single largest source of elevated values. In an electric phosphorus furnace there is no refining stage, no oxygen blow and no vacuum treatment, so whatever titanium enters with the feed can only leave in the slag, in the ferrophosphorus or in the off-gas dust. There is no operation comparable to decarburisation that a producer can switch on to pull titanium back out of a finished heat.

The chemistry explains why. Titanium dioxide is one of the most stable oxides in the burden; its reduction to metallic titanium by carbon requires temperatures well above the 1400-1500 C window in which the phosphorus reaction is run. Under normal furnace conditions only a fraction of the TiO2 is reduced, and that fraction dissolves in the phosphide melt or forms titanium phosphide and titanium carbide particles that report largely to the alloy because of its density. The remainder leaves with the slag. The distribution ratio between alloy and slag is therefore sensitive to slag basicity, to the silica-to-lime ratio, to tapping temperature and to how much slag is carried over at the tap. Two producers using the same rock can end up with different titanium in the alloy purely because of slag practice.

The practical consequence for a buyer is that titanium content is a feed-quality signature, not an adjustable dial. A producer who wants a low-titanium guarantee has to start by selecting a rock source with a low titanium-to-phosphorus ratio, by keeping titanium-bearing scrap out of the iron unit, and by controlling slag carryover so that titanium-rich slag does not contaminate the alloy during tapping. Dilution, that is blending a higher-titanium lot with a very clean lot, is the only correction available after the fact, and it is arithmetic rather than metallurgy.

Composition Windows and How to Read a Certificate

Commercial ferrophosphorus is normally sold against a table of element limits rather than against a single grade number, and the titanium line is the one that separates a standard product from a low-titanium product. The ranges below reflect what is commonly achievable in electric-furnace co-product material and are stated as typical commercial windows rather than as values from a single specification.

Product class	P (%)	Si (%)	C (%)	S (%)	Mn (%)	Ti (%)
Standard ferrophosphorus	23.0-27.0	3.0 max	1.0 max	0.5 max	2.0 max	1.0 max
Low-titanium ferrophosphorus	24.0-27.0	2.0 max	0.5 max	0.3 max	1.5 max	0.5 max
Restricted low-titanium ferrophosphorus	24.0-26.0	1.5 max	0.3 max	0.2 max	1.0 max	0.3 max
Ultra-low-titanium ferrophosphorus	24.0-26.0	1.0 max	0.2 max	0.15 max	0.8 max	0.10 max

Read the table with two cautions. First, the phosphorus figure is a range, not a guaranteed minimum alone: because the alloy is the phosphorus carrier, a shipment at 23.5% P delivers about ten percent less phosphorus per tonne than one at 26% P, and the effective cost per unit of phosphorus differs accordingly. Second, the silicon, carbon and manganese limits are not decoration. Silicon and carbon dilute the phosphorus and raise the alloy's melting range; manganese is tolerable in most steels but has to be counted in the manganese balance of the heat; sulphur adds to the sulphur already present in scrap and matters in steels with a sulphur ceiling.

Size is specified separately, most commonly 10-50 mm, with 5-30 mm and 0-10 mm fines also traded. Size matters more than it appears: undersize fines oxidise and are lost to the slag or to the dust collector when blown into a ladle, while oversize lumps dissolve slowly and can sit undissolved at the bottom of the ladle until the heat is tapped. A buyer adding the alloy through a bunker system should match the size band to the charging equipment, not simply order the cheapest band available.

How Low Can Titanium Actually Be Controlled

The answer depends on which lever the producer is allowed to pull, and every step down the ladder costs money. The table below sets out the usual routes and what each one buys.

Control route	Typical Ti ceiling	What it requires	Commercial effect
Unrestricted burden, mixed iron units	0.8-1.2%	No feed selection	Lowest cost, unsuitable where Ti is specified
Selected rock, screened iron unit	0.5-0.7%	Rock source with low Ti/P ratio; scrap exclusion; slag control	Moderate premium, widely available
Dedicated low-Ti campaign	0.2-0.3%	Segregated feed, dedicated campaign, more frequent analysis	Higher premium, limited tonnage per campaign
Blending and dilution	0.1% and below	Large stock of very clean material, homogenisation, per-lot verification	Highest cost, small lots, long lead time

Two arithmetic points are worth carrying into a negotiation. The first is dilution. If 30 tonnes of alloy at 1.2% Ti is blended with 10 tonnes at 0.2% Ti, the arithmetic mean is 0.95% Ti, and the blend is only as homogeneous as the mixing operation makes it, which is why blended lots need a composite sample rather than a grab sample. The second is the ceiling effect: guarantees below about 0.1% Ti are difficult to sustain because the analytical uncertainty and the natural lot-to-lot scatter of a co-product furnace are of the same order as the limit itself. A supplier offering 0.05% Ti without a described sampling plan is offering a number, not a control.

The realistic reading of the common question is therefore this. A conventional low-titanium ferrophosphorus holds titanium below 0.7%, and a well-run restricted grade holds it below 0.3%. Values below 0.1% are obtainable by selection and dilution but should be treated as a special order with a defined sampling and arbitration clause. Nothing in the alloy can be improved after tapping, so the guarantee is only as good as the feed segregation behind it.

What Titanium Carryover Does to the Steel

It is easy to dismiss 0.3% Ti in an additive as irrelevant. The dilution factor is large, but the addition rate is not small when the aim is a real phosphorus level. A simple rule covers most cases: titanium added to the steel in parts per million equals the ferrophosphorus addition in kilograms per tonne multiplied by the titanium percentage in the alloy, multiplied by ten. Adding 6 kg/t of a 0.7% Ti alloy therefore introduces about 42 ppm Ti, while the same addition of a 0.1% Ti alloy introduces about 6 ppm. For a 0.08% P aim with a 25% P alloy the addition is roughly 3.2 kg/t before recovery losses, and the titanium figure follows directly.

Whether 40 ppm of stray titanium matters depends entirely on the steel. In a plain carbon structural grade it is usually harmless. In boron-treated steel it is not: boron hardenability depends on boron remaining in solution, and titanium competes for nitrogen and for boron itself, so a residual that varies from heat to heat turns a predictable hardenability response into a scatter band. In titanium-stabilised and interstitial-free grades the residual disturbs the balance between titanium, carbon, nitrogen and sulphur, and the stabilisation calculation that the steelmaker has tuned no longer holds. In bearing, rail and other cleanliness-critical steels, titanium carbonitrides form hard particles that cluster, and clusters that reach the tundish nozzle are a clogging risk and an inclusion-count risk at the same time.

There is also a scrap-loop effect. Titanium added through an alloy does not disappear when the steel is recycled; it accumulates in the internal scrap stream and reappears in the next heat, so a melt shop that tolerates a loose titanium specification upstream may find itself fighting residuals in a grade where titanium is a controlled element. This is why procurement teams in shops with tight residual control tend to specify the titanium ceiling on the additive rather than accept a generic certificate.

Phosphorus itself has to be handled with equal care for the opposite reason. Phosphorus cannot be removed in conventional primary steelmaking; it is controlled by charge selection and by the amount added. That makes ferrophosphorus a one-way addition: weighing accuracy of one or two kilograms per tonne is not a formality, because an over-addition cannot be corrected by a longer blow. In steels where phosphorus is deliberately raised, such as free-machining grades, certain weathering grades and some non-oriented electrical steels where phosphorus raises resistivity and lowers core loss, the additive is a precision ingredient rather than a bulk deoxidiser.

Sampling, Analysis and Acceptance Testing

Titanium is the element most likely to be argued about at acceptance, and almost every dispute traces back to sampling rather than to the laboratory. Ferrophosphorus is a heterogeneous solidified melt: phosphide and carbide phases are not evenly distributed, and a single lump can differ from the lot average by more than the specification tolerance. Sampling and sample preparation should follow a recognised ferroalloy procedure such as ISO 3713 or GB/T 4010, with increments taken from a defined number of lumps across the lot, crushed and reduced to a laboratory sample fine enough for the analytical method used. General delivery requirements, packing and marking are commonly written around GB/T 3650, and size distribution is checked by manual sieving in accordance with the applicable ferroalloy sieving method.

On the analytical side, phosphorus is usually determined gravimetrically or spectrophotometrically after dissolution, silicon and manganese by inductively coupled plasma optical emission spectrometry or by X-ray fluorescence on fused beads, carbon and sulphur by combustion with infrared detection, and titanium by ICP-OES or by spectrophotometry. At a 0.3% limit the laboratory needs a working range that extends well below 100 ppm so that the reported figure carries a defensible uncertainty; at a 0.1% limit that requirement becomes the dominant selection criterion for the laboratory itself. A certificate that states a titanium value without stating the method is difficult to enforce.

A workable acceptance clause contains five items: the lot definition, the sampling standard and increment count, the analytical methods and the reporting resolution, the retest and arbitration route for a failed result, and the treatment of fines and moisture. The last item is often forgotten. Fines below the agreed lower size limit are not equivalent to lump on a phosphorus-per-tonne basis because they are lost during charging, and a consignment that is nominally within chemistry but heavily degraded in size can still deliver a shortfall in recovered phosphorus.

Packing, Storage and Melt-Shop Handling

Ferrophosphorus is shipped in one-tonne jumbo bags with an inner moisture barrier, in 25 kg bags on pallets, or in steel drums for smaller specialty lots. The inner liner is not cosmetic. The alloy reacts slowly with moisture, and the reaction releases phosphine, a toxic and at higher concentrations self-igniting gas; damp storage is therefore both a quality problem and a safety problem. Occupational exposure limits for phosphine are of the order of 0.3 ppm as an eight-hour time-weighted average in most jurisdictions, well below the odour threshold of many people, which is why reliance on smell is not an acceptable control.

Storage should be dry, covered and ventilated, away from acids and oxidisers, and separated from any area where water can pool or where steam lines can drip. Water should not be used on a fire involving ferrophosphorus, because the reaction with water generates more phosphine; dry powder or suitable dry media are the appropriate first response, and the area should be evacuated and monitored. In the melt shop, the alloy should be added to deoxidised steel rather than to a strongly oxidising bath, and preferably late in the tap with the addition chute positioned to let the dense lumps sink through the slag rather than ride on it. Recovery of phosphorus from a correctly sized, correctly timed addition is high, typically in the range of 85-95%, because phosphorus is not oxidised out of the bath under normal tapping conditions.

Bulk stock should be rotated on a first-in-first-out basis and lots should remain traceable to their certificate. Because the material is brittle, re-handling generates fines; each additional transfer reduces the usable lump fraction, so buyers who take jumbo bags directly to the charging bunker rather than through an intermediate re-bagging step generally see better recovery per tonne purchased.

FAQ

Q: How low can the titanium content in low-titanium ferrophosphorus be controlled?
A conventional low-titanium grade holds titanium below 0.7%, and a restricted grade produced from selected rock and screened iron units typically achieves 0.3% or lower. Values around 0.1% are possible by selection and blending, but they are special-order limits that depend on a defined sampling plan as much as on furnace practice.

Q: Does lowering the titanium limit change the phosphorus content?
Not directly, but the two are linked through feed quality. Cleaner feed usually means a narrower phosphorus band, commonly 24-26% P rather than 23-27%, so the phosphorus guarantee should be renegotiated at the same time as the titanium guarantee.

Q: How is titanium in ferrophosphorus measured?
A composite sample is taken according to a ferroalloy sampling standard such as ISO 3713 or GB/T 4010, crushed and reduced, then analysed by ICP-OES or spectrophotometry for titanium, with phosphorus, silicon, manganese, carbon and sulphur determined by their respective methods on the same sample.

Q: Why can titanium not be removed from the alloy after melting?
Titanium dioxide is far more stable than iron oxide and is only reduced at temperatures well above the operating window of the phosphorus furnace, so there is no practical refining step that strips titanium from a phosphide melt. Control has to be built into feed selection and slag practice before the tap.

Q: What happens if a 0.7% Ti alloy is used in boron-treated steel?
At an addition rate of about 6 kg/t the alloy introduces roughly 40 ppm titanium, which is enough to interfere with boron hardenability and to make the hardenability response vary between heats. Grades with controlled boron or titanium should specify a 0.3% Ti alloy or lower and should verify the residual in the first heat of each lot.

Q: Is ferrophosphorus hazardous in storage?
The alloy itself is stable when dry, but contact with moisture or acid releases phosphine, which is toxic at low concentrations. Keep it dry and ventilated, avoid water on fires involving the material, and monitor the atmosphere where bags may have absorbed moisture.