HCEM HC280D oilless vacuum pump, 24 V DC

Vacuum Filtration: How It Works, How to Set It Up and How to Choose the Vacuum Pump

HCEM Engineering Team

Vacuum filtration separates a solid from a liquid by lowering the pressure under the filter, so the atmosphere above pushes the liquid through. A pump or aspirator evacuates a sealed filter flask, and the pressure difference drives the liquid through the paper or membrane much faster than gravity, leaving a drier solid.

This guide covers the apparatus and setup, plus the numbers most lab guides skip: how much vacuum the job needs, how much pump flow a station or manifold takes, and how to keep liquid and vapor out of the pump.

What is vacuum filtration, and why is it faster than gravity filtration?

In gravity filtration, only the weight of the liquid above the filter pushes it through. In vacuum (suction) filtration the flask below the filter is pumped down, so the full difference between room and flask pressure acts across the filter. The University of York's Chemistry Teaching Labs guide (read 2026-11-06) calls it the normal way to collect a solid product. Lisa Nichols' Organic Chemistry Lab Techniques on LibreTexts (read 2026-11-06) adds that it is often done in under a minute when the joints seal well and the vacuum source is strong, and that it removes more of the remaining liquid, giving a purer solid.

The reason is Darcy's law. The ETH Zurich filtration practicum notes (read 2026-11-06) give filtrate flow per unit area as the pressure drop divided by viscosity times the resistance of medium plus cake: flow rises with pressure difference and falls as the cake thickens.

Worked example: gravity vs vacuum. A 5 cm water column gives ρgh = 1,000 kg/m³ × 9.81 m/s² × 0.05 m = 490 Pa, about 0.5 kPa. A flask at 50 kPa below atmospheric gives 50,000 Pa, about 100 times more. For a cake that does not compress, the initial flow rises by about the same factor; real cakes often compact, so the gain is smaller, but it shows why a slow gravity filtration can finish in minutes under vacuum.

Limits: strong suction can pull fine crystals through the paper (Nichols), and reduced pressure can boil a volatile filtrate (York), so the method suits collecting the solid best.

What apparatus do you need for vacuum filtration?

Part What it does What to check
Büchner funnel (porcelain, glass or plastic) or Hirsch funnel for small amounts Holds a flat filter paper over a perforated plate Paper covers every hole and lies flat
Membrane filter holder (funnel and base) or a multi-place manifold Water testing and microbiology on 0.45 µm membranes Each place has its own valve
Filter flask (Büchner flask) with side arm Collects the filtrate under vacuum Heavy-walled, rated for vacuum, no chips or scratches
Rubber adapter, cone or sleeve Airtight seal between funnel and flask Right size; no cracks
Trap flask Catches liquid and foam before they reach the vacuum line About the same size as the filter flask
Thick-walled vacuum tubing Connects flask, trap and vacuum source Will not collapse under vacuum; not kinked
Gauge and bleed (needle) valve Show and limit the vacuum Needed when a method caps the vacuum

For water testing, EPA Method 1604 (coliforms and E. coli by membrane filtration, read 2026-11-06) lists a 1 liter filter flask, an optional manifold, a safety trap flask and a vacuum source with a check valve.

How do you set up a vacuum filtration apparatus step by step?

  1. Inspect the glass for chips, cracks and scratches, and use only flasks made for vacuum.
  2. Clamp the filter flask at the neck.
  3. Connect side arm to trap, and trap to vacuum source, with short thick-walled tubing; Nichols notes that bent tubing gives poor suction. Fit the adapter and funnel.
  4. Place the filter so it covers every hole and lies flat; Nichols puts the paper's slight curve facing down so solid cannot creep under the edge.
  5. Wet the filter and start the vacuum. If the solvent does not drain, press the funnel down to seat the seal.
  6. Pour into the center in portions, swirling first, and rinse the flask with a little cold solvent.
  7. Wash the cake if needed: break the vacuum, add a little cold solvent, stir gently, reapply the vacuum.
  8. Vent first, then stop the vacuum source. Nichols explains that turning off an aspirator while the flask is still under vacuum can pull sink water back into it.
  9. Watch the level: York says to stop and empty the flask before the filtrate approaches the side arm.

How much vacuum does vacuum filtration need?

Much less than most pumps make. Filtration is rough vacuum work, and methods set a working range or a ceiling, not the deepest vacuum the pump can reach:

  • Membrane filtration for coliform bacteria. Standard Methods 9222 (approved 2006; copy in the EPA docket, read 2026-11-06) asks for a pressure differential of 34 to 51 kPa across the membrane.
  • Chlorophyll filtration. EPA Method 445.0 (read 2026-11-06) says vacuum filtration should not exceed 6 inHg (20 kPa), because higher pressures or filtrations longer than about 10 minutes can damage the algal cells, and it tells you not to suck the filter dry.
  • Büchner filtration of crystals. No fixed figure. Stronger vacuum is faster until the paper tears, fine crystals pass through, or the filtrate starts to boil. York suggests a gentle "passive" vacuum (close the valve to the vacuum source, then switch the source off, so the sealed flask keeps its reduced pressure) for filtrates in volatile solvents.

Gauge readings (kPa or inHg below atmosphere) depend on local air pressure; absolute pressure (mbar) is measured from zero. The table converts for a standard atmosphere of 101.325 kPa, using the NIST Guide to the SI factors (read 2026-11-06): 1 inHg = 3.386 kPa, 1 mbar = 0.1 kPa.

Gauge (kPa) Vacuum (inHg below atmosphere) Absolute (mbar) Where it comes up
−20 kPa 5.9 inHg 813 mbar EPA 445.0 maximum for chlorophyll filtration
−34 kPa 10.0 inHg 673 mbar Lower end of the Standard Methods 9222 range
−51 kPa 15.1 inHg 503 mbar Upper end of the Standard Methods 9222 range
−80 kPa 23.6 inHg 213 mbar Strong vacuum; low-boiling solvent filtrates may boil (York)
−92 kPa 27.2 inHg 93 mbar HCEM HC280D/HC480/HC580A open-flow rating
−98 kPa 28.9 inHg 33 mbar Deepest rating of the four HCEM models below

So a pump rated at −90 kPa or better has plenty of vacuum; add a gauge and a bleed valve to set the level a method allows. A water filtrate boils only near full vacuum (water's vapor pressure is 17.5 mmHg at 20 °C, about 23 mbar, as Nichols notes); low-boiling organic solvents boil at far milder vacuum, as York warns. More on the scales in gauge vs absolute pressure.

How much pump flow does a filtration station or manifold need?

Little for pumping down; more for the air that keeps coming in. Pump-down time is t = (V ÷ S) × ln(p0 ÷ p1): V the volume, S the average effective pump speed, p0 and p1 the start and target absolute pressures (see our evacuation time guide).

Worked example 1: one Büchner station. Flask, trap and tubing of 2.1 L to 500 mbar (about −51 kPa) at 20 L/min: t = (2.1 ÷ 20) × ln(1013 ÷ 500) = 0.105 × 0.71 = 0.074 min, about 4.5 seconds.

Worked example 2: a six-place membrane manifold. Standard Methods 9222 mentions manifolds of three to six filter assemblies. With V = 3.4 L (manifold, 2 L flask, 1 L trap, tubing), pump-down to 500 mbar at 20 L/min takes (3.4 ÷ 20) × 0.71 = 0.12 min, about 7 seconds. The real load comes when a funnel is lifted or a place is left without a membrane with its valve open, or a paper cake cracks, and room air flows straight in. Assume, for illustration, 5 L/min of free air (measured at atmospheric pressure) per open place, two places open at once, and 1 L/min of joint leakage: 11 L/min of free air. At 500 mbar each liter of free air fills 1013 ÷ 500 = 2.03 liters, so to hold 500 mbar the pump must move 11 × 2.03 = 22 L/min (0.79 CFM) at that pressure.

The 5 L/min is an assumption; measure your own setup. The method is the point:

  • Size on the pump's speed at your working pressure, not the datasheet's free-air figure; speed falls toward zero at the pump's ultimate vacuum, as ultimate pressure vs working vacuum explains.
  • Give every manifold place its own valve, and close it as soon as that place's filtration ends or its funnel comes off. That cuts the air load more than a bigger pump does.
  • Add margin for ageing seals; our CFM to L/min converter handles mixed units.

Which vacuum source is best for filtration?

Source Vacuum it reaches Strengths Trade-offs
Water aspirator No lower than water's vapor pressure, 17.5 to 23.8 mmHg at 20 to 25 °C (Nichols); worse when flow drops Cheap, no motor Uses a lot of water; vacuum varies with flow; back-suction risk
Diaphragm pump Depends on the number of stages Oil-free Compare with piston pumps in diaphragm vs piston pumps
Oil-free piston pump The HCEM models below are listed at −90 to −98 kPa gauge Oil-free, no water Built to move air; protect it from liquid and condensable vapor
Oil-sealed rotary vane pump Far deeper than filtration needs Deep vacuum Solvent vapor contaminates the oil; oil changes and waste oil; needs a cold trap
House (central) vacuum Depends on the building system Always on, no pump on the bench Shared with other labs; vapors go into the building system

UAB Sustainability (read 2026-11-06) reported its chemistry-lab aspirators using nearly four gallons of water per minute, 40 hours a week, before replacing them with vacuum pumps, and the EPA's WaterSense at Work guide (read 2026-11-06) notes that aspirator water goes once to the drain while an air-cooled dry pump uses none. Nichols warns that solvent from filter flasks can degrade the oil pumps of a house vacuum, and Yale EHS (read 2026-11-06) that an in-line HEPA filter does not stop solvent vapor entering it.

How do you protect the pump and work safely?

  • Always fit a trap. Duke OESO and Washington State University EHS (both read 2026-11-06) put a trap between apparatus and vacuum source first: it keeps liquid out of the pump and lines and vapor out of the room. WSU suggests a second trap bottle on filtration setups to catch overflow from the first, and Standard Methods 9222 asks for a trap of about the filter flask's capacity. York notes that a pump loses much of its effectiveness if anything but gas is drawn into its inlet.
  • Condense vapor before the pump. With a volatile solvent, a cold trap in front of the pump catches what evaporates; using only the vacuum the filtration needs reduces it. Pump inlet filters are a separate subject: see our inlet filtration guide.
  • Use glass made for vacuum. Duke OESO warns that ordinary glassware can implode and calls for heavy-walled filter flasks; uncoated glass filter flasks must comply with ASTM E1406, and PVC-coated or plastic flasks do not throw shards. Even a filter flask can implode, so it must be taped or shielded, and a face shield should be worn. WSU says never to evacuate an ordinary flat-bottomed flask.
  • Inspect. Yale EHS recommends filter flasks under 4 liters, checked for cracks, chips, etching or scoring at every cleaning; in one of its incidents a 4-liter flask not rated for vacuum imploded during filtration.
  • Vent before you stop. Never stop the vacuum source while it is still open to the evacuated flask. Open the system to air first (Nichols), or close the valve between the apparatus and the pump before switching the pump off (WSU), so water, filtrate or pump oil cannot be drawn back.
  • Use a fume hood for hazardous liquids, and avoid aspirating toxic, volatile or corrosive liquids (Duke).

Why is my vacuum filtration slow?

Symptom Likely cause What to do
No suction at the funnel Leak at the adapter, loose tubing or a weak source Press the funnel down to seat it; test the source by covering the tubing end with a finger (Nichols); check the gauge
Fast at first, then slow Cake or fine particles blocking the filter Normal by Darcy's law as the cake grows; let solids settle and pour the liquid first, or use a filter aid such as Celite (York)
Manifold vacuum falls when one place is opened Air flowing through the open place Close that place's valve; size the pump as in worked example 2
Pump reaches less vacuum than it used to Worn seals or valves, or liquid in the pump Check the trap, then test the pump against its rating: see how to test a vacuum pump

Where does an oil-free piston pump fit, and which HCEM models suit filtration?

An oil-free piston pump suits filtration benches, membrane manifolds and instruments that filter in their cycle: no oil, no water, and more vacuum than filtration needs. Do not assume chemical resistance: tell us your solvents and we will confirm whether a piston pump suits them, and fit a trap (and a cold trap where needed) in any case. Evaporation and drying are covered in our oil-free lab vacuum pump article and the vacuum oven pump guide.

These four pumps from our oil-free vacuum pump range list laboratory filtration among their applications. Each parameter table gives an open-air flow and a flow at maximum vacuum, quoted as listed:

Model Supply Parameter table: open-air flow / flow at max. vacuum Power Noise Weight
HC100A (HC100B) 220 V/50 Hz (110 V/60 Hz) 45 L/min (1.59 cfm) open flow (max. −90 kPa) or 22.5 L/min (0.80 cfm) (max. −98 kPa); HC100B: 50 L/min (1.77 cfm) or 25 L/min (0.88 cfm) ≤110 W (≤150 W) ≤50 dB(A) (≤51 dB(A)) 3.4 kg
HC280D (HC280C, HC280E) 24 V DC (12 V, 48 V DC) 90 L/min (3.18 cfm) open flow (max. −92 kPa) or 45 L/min (1.59 cfm) (max. −98 kPa) ≤180 W ≤58 dB(A) 4.9 kg
HC480 (HC480A, HC480B) 220 V/50 Hz or 110 V/60 Hz 95 L/min (3.35 cfm) open flow (max. −92 kPa) or 47.5 L/min (1.68 cfm) (max. −98 kPa); HC480B: 100 L/min (3.53 cfm) or 50 L/min (1.77 cfm) ≤200 W or ≤260 W ≤53 or ≤55 dB(A) 9.8 kg
HC580A (HC580B) 220 V/50 Hz (110 V/60 Hz) 120 L/min (4.24 cfm) open flow (max. −92 kPa) or 60 L/min (2.12 cfm) (max. −98 kPa) ≤340 W ≤60 dB(A) (≤61 dB(A)) 10.1 kg

Figures are from each product's parameter table (store data of 2026-10-31). Sizing as in worked example 2 needs the flow at your working pressure, which the table does not give: ask us before you design a pump in. All four reach deeper than the Standard Methods 9222 range or the EPA 445.0 limit, so pair them with a gauge and a bleed or regulating valve.

For equipment makers we offer custom voltages and logos from 50 units; qualified OEM customers and distributors can apply for samples (subject to review); samples take 10 to 20 days. See the OEM and ODM page.

Common questions

Can I use a vacuum pump instead of a water aspirator?

Yes. York treats both as adequate for routine work; a pump's vacuum does not depend on the water supply. Keep a trap in front of either.

What vacuum level should I use for membrane filtration?

Follow your method. For coliform testing, Standard Methods 9222 gives a 34 to 51 kPa pressure differential (10 to 15 inHg); for chlorophyll, EPA 445.0 limits it to 20 kPa (6 inHg).

Choosing a pump for a filtration bench, manifold or instrument? Send us the number of places, the vacuum your method allows, the liquids and solvents, and your supply voltage. Our engineers will suggest a model, tell you what flow data we have at your working pressure, and tell you plainly if a piston pump is not the right fit.

Sources

All read 2026-11-06.

Back to blog