[theThinker (OP)]

I'll like to know how accurate are our chemical balances.

e.g. What is the highest accuracy when weighing to 1 gram.

[syhprum]

Weight equates to the Earths gravitational attraction to a given mass and varies according to where you make the measurement due to variations in the underlying structure of the Earth ,centripetal effects due to its rotation and the buoyancy of the atmosphere.
A perfectly accurate weighing machine could given a variation of reading of at least 0.1% due to these effects.
if you are comparing the mass of a sample by balancing it against standard masses many of these effects cancel out with only the buoyance the atmosphere producing and error if the specific gravity of the sample is different to that of the standard masses and if the standard masses have been properly corrected for buoyance effects. 

[alancalverd]

In principle, the achieveable precision of an evacuable balance is limited by the rate of evaporation or radioactive decay of the sample. In practice, microgram resolution of laboratory samples (i.e. about 1 part in 10⁹) is usually adequate for the task, but the systematic uncertainty of the International Prototype Kilogram exceeds the sensitivity of the best balances, so whilst you might compare two masses to better than 1 in 10⁹, you wouldn't know them to better than 1 in 10⁷. Hence the reason for the recent redefinition of the kilogram. 

Specifically, I'd expect to routinely resolve a gram of reasonably dense material to ± 10μg with professional laboratory equipment. To do better (~1 g of pyrolytic graphite ± 0.5 μg) I've used an evacuated  Cahn Electrobalance, but here we are talking about evaporation limits and investigative physics and engineering rather than everyday chemistry.

[evan_au]

What is the highest accuracy when weighing to 1 gram

A couple of comments about terminology:
- As described by Syphrum, the weight of an object is affected by many variables, including the weather and barometric pressure. So physicists prefer to measure an object's mass.
- The pound is a unit of weight, while a kilogram is a unit of mass.
- Given an international standard kilogram, national measurement laboratories can accurately measure masses of 1 gram, 1 milligram, 1 microgram, 1 tonne, etc.
- There has been a "contest" over the past 5 years to redefine the international standard kilogram so that it wasn't defined by a lump of metal carefully locked away near Paris.
- Last year, the contest was won by the Kibble balance, based on Planck's constant. Plank's constant is now defined to 10 decimal places.
- However, the "runner-up" in this contest gave us another extremely accurate way to measure matter - by actually counting the atoms in a 1kg sphere of isotopically pure silicon. This allowed a much more accurate definition of Avogadro's number.
- So in theory, you could measure the mass of a quantity by counting the atoms...

In reality, the achievable accuracy of measuring mass depends on what you are trying to achieve, how much you are willing to pay, and how long you are willing to wait.
- There are local regulations that govern the accuracy of supermarket scales. These can be calibrated easily and cheaply.
- There will be tighter regulations on the calibration of devices that are used for calibrating supermarket scales...
- There are probably no legal constraints on the accuracy of your bathroom scales.

But if you want to measure mass really accurately, you will do it in a vacuum chamber.
- If your sample would misbehave in a vacuum chamber (eg outgassing or suffocation), then you will need to contain it in a sealed capsule (which you will also need to measure while it is empty).

See: https://www.nist.gov/si-redefinition/kilogram-silicon-spheres-and-international-avogadro-project

Oops - crossover with alancalverd

[Bored chemist]

I'll like to know how accurate are our chemical balances.

e.g. What is the highest accuracy when weighing to 1 gram.

It depends on the operator.
Not least, most people don't even know about the effects of air buoyancy so their measurements might easily be out by about 0.1%- even though they slavishly write down all 6 or 7 digits that it says on the display.

An analytical balance with a stable, relatively low density, test mass ( something like a glass stopper) makes a good enough barometer for tracking the weather.
A bit of googling suggests that you can measure 1 gram to about 0.15µg - roughly 1 part in ten million. How good do you need?

There are also interesting problems with calibrating some balances.
To calibrate the balance, you need test masses with masses known to better precision hat the  resolution of the balance.
But you can't get a better estimate of the mass than by using the best balance.

[theThinker (OP)]

Of course, we assume the experimenter knows about the effect of air buoyancy, etc.. I am talking about measuring mass. I believe the user of a certain type of balance should know about gravitational effectstoo.

So it seems we may get to about 1 part per million or more in very professional laboratories. This is what I want to know.

Thanks.

[Bored chemist]

Of course, we assume the experimenter knows about the effect of air buoyancy, etc.

I wouldn't assume that.

[ConnorGunning]

Of course, we assume the experimenter knows about the effect of air buoyancy, etc.. I am talking about measuring mass. I believe the user of a certain type of balance should know about gravitational effectstoo.

So it seems we may get to about 1 part per million or more in very professional laboratories. This is what I want to know.

Thanks.

There's levels of precision of course, don't forget about that.

[sxgrt]

In principle, the achieveable precision of an evacuable balance is limited by the rate of evaporation or radioactive decay of the sample. In practice, microgram resolution of laboratory samples (i.e. about 1 part in 10⁹) is usually adequate for the task, but the systematic uncertainty of the International Prototype Kilogram exceeds the sensitivity of the best balances, so whilst you might compare two masses to better than 1 in 10⁹, you wouldn't know them to better than 1 in 10⁷. Hence the reason for the recent redefinition of the kilogram. 

Specifically, I'd expect to routinely resolve a gram of reasonably dense material to ± 10μg with professional laboratory equipment. To do better (~1 g of pyrolytic graphite ± 0.5 μg) I've used an evacuated  Cahn Electrobalance, but here we are talking about evaporation limits and investigative physics and engineering rather than everyday chemistry.

Hello. I just want to know. When the number on the scale shifts between, say, 8.9955 and 8.9956 without settling on one or the other, which do I use?

[alancalverd]

8.99555 ± 0.00005

[Bored chemist]

In principle, the achieveable precision of an evacuable balance is limited by the rate of evaporation or radioactive decay of the sample. In practice, microgram resolution of laboratory samples (i.e. about 1 part in 10⁹) is usually adequate for the task, but the systematic uncertainty of the International Prototype Kilogram exceeds the sensitivity of the best balances, so whilst you might compare two masses to better than 1 in 10⁹, you wouldn't know them to better than 1 in 10⁷. Hence the reason for the recent redefinition of the kilogram. 

Specifically, I'd expect to routinely resolve a gram of reasonably dense material to ± 10μg with professional laboratory equipment. To do better (~1 g of pyrolytic graphite ± 0.5 μg) I've used an evacuated  Cahn Electrobalance, but here we are talking about evaporation limits and investigative physics and engineering rather than everyday chemistry.

Hello. I just want to know. When the number on the scale shifts between, say, 8.9955 and 8.9956 without settling on one or the other, which do I use?

Either.
If it matters, you are not using a good enough balance.
Also, how are you measuring the air density?