Wet bench design for a fabrication laboratory in Bangladesh
The chemistry list decides the bench, not the other way round. Materials, extraction and duct routes, acid and solvent segregation, spill containment, heated baths, DI water and what has to happen before anything reaches a drain.
A wet bench is delivered to a new fabrication laboratory in Dhaka with a 200 mm extract spigot on the back. The building riser is 150 mm, it serves three other rooms, and the fan on the roof was sized for a chemistry store. The bench sits against a wall for six months while the mechanical works are retendered.
Wet processing is easy to underestimate and hard to retrofit. The bench is not complicated. The services around it, and the decisions about what chemistry will be used, are where the engineering sits.
Decide the chemistry list before the bench
Every material, ventilation and safety decision follows from the list of chemicals the laboratory intends to use. Write that list first, including what a future project might need, and be honest about it.
Hydrofluoric acid changes everything. It attacks glass and silica, so no glassware and no glass-lined anything. It requires calcium gluconate gel on site with an in-date stock, specific training, a two-person rule and a documented first aid procedure that the nearest hospital has been told about. If HF is on the list, that decision drives the whole bench specification.
Piranha is not storable. Sulphuric acid with hydrogen peroxide heats itself, evolves gas and continues reacting. It cannot be sealed in a container, it must be used and then cooled and neutralised, and a bottle of used piranha in a waste cabinet is a genuine hazard.
Hot phosphoric acid for nitride etching runs near its boiling point and needs a reflux condenser to hold the water concentration, which fixes the etch rate. Without the condenser the bath drifts continuously.
Perchloric acid needs its own bench and a washdown duct. Perchlorate deposits in ordinary ductwork become explosive over time. If anybody plans perchloric work it shares nothing with anything else, and that has to be settled before a single bench is quoted.
Bench material and what it survives
Polypropylene is the standard for acid work. It is chemically excellent, easy to weld, and inexpensive. Its weaknesses are temperature and fire behaviour, so specify a flame retardant grade and keep hot work away from it.
PVDF tolerates higher temperature and behaves better in a fire, and it is what heated acid sections are usually made from. It costs considerably more, which is why a mixed construction with PVDF only where it is needed is common.
Stainless steel suits solvent benches and does not suit acids. A stainless bench used for HF is a short-lived bench.
Viewing panels need checking against the same chemistry list. Polycarbonate is attacked by solvents and some bases, and a clouded panel that cracks is a containment failure. Specify white work surfaces while you are at it: staff clean what they can see.
Extraction, duct material and the stack
The bench captures fume at the source through a rear plenum with a slotted baffle, and the whole design depends on that airflow being right and staying right.
Face velocity is the specification, measured across the working aperture at the sash height staff will actually use, at several points, and recorded at commissioning. A single reading in the middle proves nothing.
Make-up air has to come from somewhere. In a small university cleanroom, a wet bench extract can be a large fraction of the total supply. Without matched make-up air the room loses its pressure cascade, the cleanroom becomes negative to the corridor and the class is gone. This is the most common design error in a converted room.
Duct material follows the chemistry. Polypropylene or CPVC for acid, stainless for solvent. Galvanised duct on an acid bench corrodes from the inside and fails without warning, and acid and solvent streams should not share a duct without a specific risk assessment.
Airflow monitoring at the bench with an alarm, not a flap on a string. Staff need to know the extract has failed before they open a container.
The stack terminates above roof level, discharging upwards at velocity, away from your own fresh air intake and from the neighbouring building's. On a dense Dhaka campus the neighbour's intake is often the closer one, and it is worth walking the roof before the drawings are approved.
Segregating acids, solvents and bases
Segregation applies at the bench, in the store and in the waste stream, and it is a physical separation rather than a labelling exercise.
Separate under-bench cabinets for acids, bases and solvents, each with its own containment and ventilation.
Solvents in a ventilated flammables cabinet, earthed and bonded for dispensing from a large container. Static from pouring solvent is a real ignition source and is routinely ignored.
Oxidisers away from organics. Nitric acid and solvent waste in one container is among the few laboratory mistakes that reliably produce an explosion.
Waste containers segregated at the point of use. Once mixed, waste costs far more to dispose of and often cannot be characterised at all.
Spill containment and the plumbing underneath
What happens under the bench decides how bad a spill gets.
Bund the bench itself, sized for the largest single container in use, with no penetration that leaks it into the floor void.
No open floor drain in the chemical area unless it goes to neutralisation. A drain running to the campus sewer is a route for a spill to leave the building unrecorded.
Drip trays under every valve and union, and double containment on any line carrying chemistry through a wall or ceiling void.
Valve and fitting materials matched to the chemistry. No brass, copper or aluminium anywhere near hydrofluoric acid, and a labelled isolation valve a student can find without opening a cabinet.
Spill kits sized for the largest container in the room, acid and solvent kept separate, checked and restocked on a schedule rather than after they are used.
Heated baths and quartz
Heated chemistry is where wet benches injure people, and the protections are specific.
An over-temperature cut-out independent of the controller. If the control thermocouple fails, the controller calls for full power indefinitely. The cut-out is a separate device with its own sensor and it is not optional.
A low level cut-out. A heater running dry cracks a quartz tank and can start a fire. Level sensing is cheap and prevents the most common heated bath failure.
Controlled ramp rates on quartz. Thermal shock cracks quartz, and a cracked quartz tank full of hot acid is the worst outcome on a wet bench.
Reflux condensers where the chemistry needs a fixed composition, with cooling water that is interlocked so the heater cannot run without flow.
Timers with audible alarms. Long etches get forgotten, and an unattended hot bath is how a process becomes an incident.
Rinsing and DI water
Deionised water is a process chemical, and in Bangladesh it needs more attention than it usually gets.
Specify resistivity, total organic carbon and particle count at the point of use, not at the outlet of the polisher. What matters is what reaches the substrate after the pipework.
Recirculate, and design out dead legs. Stagnant DI water grows bacteria fast at Dhaka temperatures, and a dead leg recontaminates the loop every time it is disturbed. Ultraviolet treatment in the loop and a point-of-use filter are the usual controls.
Match the pretreatment to the incoming supply. Municipal and deep tube well water differ in hardness and dissolved iron, and a polisher sized on a generic assumption consumes cartridges far faster than budgeted. Have the feed water analysed first.
Choose the rinse method deliberately. Overflow rinsing uses a great deal of water, quick dump rinsing less, and a spin rinse dryer handles small batches with least. In a building on a rooftop tank, water per run is a real constraint.
Hold spare cartridges and membranes locally. A DI plant waiting on an imported cartridge stops the whole wet area.
Neutralisation and what goes to the drain
Nothing from a fabrication wet bench should reach a campus drain untreated, and the Department of Environment will ask what leaves the building.
A pH neutralisation tank with dosing and monitoring on the acid and alkali drain line, with the pH logged before discharge. A dosing system with no record of what it discharged is difficult to defend.
Fluoride needs a separate step. Neutralising HF waste to a safe pH does not remove fluoride. Precipitation with a calcium salt and removal of the resulting solid is the normal route, and the solid is then a waste that has to be disposed of.
Solvents never go to drain. They are collected, segregated and handed to a licensed disposal contractor. In Bangladesh that arrangement has to be made in advance, with the contractor identified and the documentation agreed, because it is not something you can organise on the day a drum fills.
Label and log every waste container with contents and start date. Unlabelled waste is the most expensive item a laboratory can produce, because it has to be characterised before anybody will take it.
Design the waste route into the building: where drums are stored, how they leave, and whether a vehicle can reach that point. A waste store on the fourth floor behind a weight-limited lift is a design that fails quietly.
Emergency equipment, and the checks that get skipped
Eyewash and safety shower within a few seconds of the bench, on the same level, with an unobstructed path. Measure that path with the bench occupied, not on a drawing.
Tempered water to the shower. A safety shower fed from a rooftop tank in June delivers water hot enough to make a casualty step out of it. Tempering valves are cheap and this failure is specific to buildings like ours.
A floor drain or a plan for the water. A safety shower activated in a cleanroom without a drain floods the facility. That is an acceptable trade, but the plan for the aftermath should exist before it happens.
Weekly flush and log, because an eyewash unused for months delivers whatever grew in the stagnant leg into an injured eye.
Calcium gluconate gel with a checked expiry date wherever HF is used, at the bench rather than in a cupboard down the corridor, and a two-person rule enforced through the booking system so nobody can reserve an HF session alone.
An emergency plan the local hospital knows about. HF exposure needs specific treatment, and the time to explain that is not while somebody is being driven there.