Beam break across 15m. Using reflector panel

I'm really unsure how to go forward from here.

I want to make a beam break detector which triggers a relay when the beam has been broken for 5 seconds.
I have a 10x10 cm retro reflector panel I want to reflect the beam from. This returns the beam perfectly although "diffuses" it a bit (it returns a kind of pattern instead of just a single beam).

I've searched around a lot trying to find solutions for this setup in general but i'm left with some dead ends and some questions.
If you can point me in any direction I will appreciate it.

My main concerns right now are:

Should I use an IR laser (dangerous) or red/green laser?
Should I use the Arduino to modulate the laser to 38Khz or a 555 timer?
Should I use a TSOP382 IR receiver or can I use a photoresistor to look for the modulated beam and not just look for light intensity?

I'm planning on using an Arduino Uno r3 because I have one. But I'm up for some shopping if needed.

For that distance it will be necessary to modulate the transmitted beam (38 kHz is popular) and use a photodiode receiver tuned to the same frequency. There are plenty of tutorials on line -- search for "38 kHz light barrier".

One or more narrow-beam IR LEDs (<=10 degrees) and a TSSP4038 (not TSOP!!!) might be able to do that distance.
Two units (transmitter <> receiver) would be much easier though.

Generating a continuous 38kHz signal can be done in setup.
Use a LED driver with current limiting.
Leo..

const byte IR_LED = 11; // pin 11

void setup() {
  pinMode (IR_LED, OUTPUT);
  TCCR2A = _BV (COM2A0) | _BV(WGM21);
  TCCR2B = _BV (CS20);
  OCR2A =  209; // ~209 = ~38kHz | ~219 = ~36kHz
}

void loop() {
}

Edit: I currently use the SFH4544 and the TSSP4038 (both from RS components).

semiconductor lasers are trickier to bias properly and easier to burn out, a standard IR LED and plastic lens
assembly is the standard way, with a photodiode/lens as receiver. And modulation and IR filter to
reduce stray visible light..

Thanks for the inputs.
I have a sender->receiver setup now (off the shelf from ebay) but going towards the darker time of the year I'm not sure the solar panel will keep it alive all day, hence the reflector solution.

Several of you mention the narrow IR led. Do you think it is narrow enough to be reflected from the reflector panel and still be detectable by the receiver?

What kind of "plastic lense" should I look for for the receiver and sender?

Did you measure transmitter current draw, and calculate solar and battery.
Upgrading that (if needed) could be a lot easier.

A reflection sensor is much harder than a separate transmitter and receiver.

I think the receiver, with an opening angle of 45 degrees, can benefit the most from a lense.
No experience with IR lenses, but I see borosilicate glass also can be used for 940nm.
Search for "plano-convex lense" on ebay, and Google for borosilicate glass wavelength transparency graphs.
Leo..