Hacker News new | past | comments | ask | show | jobs | submit login
Low-background Steel (wikipedia.org)
30 points by spking on Oct 25, 2013 | hide | past | favorite | 13 comments

The facts in this article are somewhat incongruous:

Radiation levels due to nuclear testing were elevated 7% over normal

The major source of radioactivity in steel is cobalt 60, which has a half life of 5.27 years

In which case one could just wait a year and the radioactivity of your steel would drop by 7%, making up for the effects of nuclear testing contamination. Put another way, steel from 1944 has been around for some 10 half-lifes of cobalt-60, meaning it has 1/2^10th as much 60Co radiation as when it was made. Why would it matter if the radioactivity was 1/2^10th or 1.07/2^10th as much as the background radiation?

I'm sure there are other isotopes which make this more of a problem, but the facts as presented in this article don't make much sense.

That's why low background steel isn't as much in demand as it used to be. It helps that it's easier to make it from scratch by filtering the air better.

Even so, there are still some rare cases where it's easier to use older steel to ensure that there is much lower background radiation. Co-60 isn't the only source of radiation, and even 1/1,000th of peak levels can still be fairly high for applications like neutrino detectors.

That's interesting and all, but why do we get a mildly-interesting wikipedia article every couple of days without any context or commentary?

I've half a mind to write a bot that submits a random article every 48 hours.

This particular one was taken from Reddit: http://www.reddit.com/r/todayilearned/comments/1p56pz/til_st...

Let's just blame it on reddit

Could someone clarify how radiation dosing and dose rates are measured?

The article says background radiation levels peaked at .15 mSv in 1963. Looking at the wikipedia page on Seiverts, I am trying to compare this to other radiation examples, but not sure how to draw a comparison.

Would a human standing outside be receiving .15 mSv per hour? year? total?

Radiation units are a bit weird. Sieverts are normalized to biological damage, because units of energy are insufficient to differentiate between 1 J of sunlight and 1 J of x-rays.

I think the 0.15 mSv the article mentions is per year. The next sentence says per year explicitly.

It's an odd reference in the context of reduced need for special pre-war II non-irradiated steel.

From the article if looks like the most important factor for that would be atmospheric concentration of Cobalt-60 (in ppm or ppb).

There are four "[citation needed]"s in the first paragraph. Surely there must be some substance out there? I did some quick Googling and couldn't find much beyond what the article already has.

The first source from the article has a list of sources of its own:


If world-war-2 steel still emits too much radiation for your application, maybe you'll want to shield your experiment with 2000 year old lead bricks from a sunken roman vessel?


What a mess we've made of this planet over just 100 years.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact