(The opinions expressed here are those of the author, a columnist for Reuters.)
By Andy Home
LONDON, May 29 (Reuters) - It's hard not to have a grudging respect for Brazil's Vale when it comes to the company's Goro nickel project in New Caledonia.
Others would surely have walked away from what must be one of the most problematic start-ups in the history of base metals.
Over-budget and years late even before the plant was first switched on in 2010, it has since been plagued by technical set-backs, unscheduled closures and, now, violent attack by locals.
Vale has remained commendably undaunted throughout.
Yet each new start has swiftly been followed by new adversity. Even rebranding the operation as Vale New Caledonia (VNC), that tried-and-tested corporate exercise in drawing a line under historical problems, hasn't worked.
Goro, using the relatively new high-pressure-acid-leach (HPAL) technology, continues to defy Vale's boundless optimism and to drain money from its bottom line.
Until Jan. 12 this year it was, however, just Vale's problem. But in the wake of the Indonesian nickel ore ban that kicked in on that date, Goro risks becoming a bigger problem for the entire nickel market.
Vale inherited the Goro project when it bought Canadian nickel producer Inco back in 2006.
It was originally slated to come into production by the end of 2008 but it was a full two years later before the plant was fired up. By which stage the capital costs had ballooned from $3.2 billion to $4.3 billion.
The problems started almost immediately, an acid spill halting commissioning in April 2010.
It was only in Q1 2011 that Vale booked the first finished product against Goro, which is designed to produce nickel oxide for further refining into utility nickel at Vale's refinery in Dalian in China.
Since then there have been repeated technical issues with the plant, particularly the acid circuit, most seriously in 2012 when the whole facility was out of action for the best part of six months.
Originally scheduled to ramp up to full 60,000-tonne per year capacity over a three-year period, Goro has in fact generated just under 30,000 tonnes of finished product in the three years since first production.
And now it's closed again. Another acid leak and a more serious one since this time it was not contained within the plant but rather contaminated a local river.
Which triggered the subsequent arson attack on the plant by local Melanesian youths, causing damage estimated at $20-30 million.
The situation appears to have calmed down, thanks to a heavy police presence around the site, and Vale is now in dialogue with the local authorities about the conditions for a restart.
Already, however, it's clear that Vale's production guidance of 40,000 tonnes for this year is not going to be met. Its previous optimism that Goro would finally break even at an EBITDA level this year also looks highly unlikely.
Initially, Goro's many teething problems were viewed by analysts as symptomatic of the challenges with the new HPAL technology.
HPAL was designed to extract nickel from laterite deposits as opposed to sulphide deposits, historically the backbone of global nickel production but a steadily depleting source of the alloying metal.
There had been a precedent for Goro's problems. The Murrin Murrin plant in Australia, now owned by Glencore, struggled for years with a litany of plant failures and shutdowns.
But Murrin Murrin finally appears to have come to grips with its misbehaving plant. Production last year of 35,900 tonnes refined nickel was the best outturn yet, albeit still short of the plant's nameplate capacity of 40,000 tonnes per year.
More telling, perhaps, is the relative success of other new HPAL projects. All have experienced start-up glitches in one form or another but nothing on the scale or duration of those at Goro.
Indeed, the Ambatovy project in Madagascar, which is the closest in terms of size and ambition, formally declared it had achieved commercial production, defined as 70-percent capacity throughput over a 30-day period, in January this year.
Vale's Goro is starting increasingly to look like an HPAL outlier, its problems site-specific rather than technology-generic.
Up until January Vale's heroic struggles trying to get Goro to live up to its potential as one of the world's largest nickel mines weren't particularly interesting to the rest of the nickel market.
Indeed, such was the obvious level of oversupply in the market in the form of all the unwanted metal sitting in London Metal Exchange (LME) warehouses, that Goro not producing was generally deemed to be a good thing.
All that changed when the Indonesian government surprised everyone by enforcing a complete ban on the export of unprocessed minerals, including nickel ore.
With the stroke of a presidential pen, the flow of material to feed China's giant nickel pig iron (NPI) sector was halted.
Lacking sufficient alternative supplies from other ore producers such as the Philippines and with multiple uncertainties about the feasibility and timing of new NPI plants in Indonesia itself, the impact on China's NPI production is expected to be huge.
To quote Macquarie Bank, which has been one of the more bullish of voices in the market after the Indonesian ban came into force, "even with aggressive rebalancing assumptions, we are on the verge of a multi-year deficit market."
And in a deficit market every tonne counts.
Including those from Goro.
Not this year. Vale's 40,000-tonne target was always ambitious even before the latest closure. Few really expected it to get anywhere close to that level.
Not maybe even next year, given the size of the stocks buffer sitting in LME sheds.
But at some stage the nickel market will need Goro's production. It just has to hope that Vale's New Caledonian nightmare is over by then. (Editing by Susan Thomas)