Blend is worthless for thermal testing. It does not apply a steady state load and primarily tests memory stability. Only run Small FFT for testing thermal and stability compliance.
Instant 100°C temperatures when you start Prime would indicate that either the water block/cold plate assembly is not seated correctly or does not have the necessary mounting pressure, OR the pump is not working.
Make absolutely certain that the pump USB and CPU_FAN connections are plugged in and plugged to the correct headers.
Most of these units come with a variety of different standoffs for mounting the pump or cold plate to the CPU socket. Make sure you used the correct ones. Using standoffs that are too tall would cause the cold plate to not be in contact with the CPU lid, and thus you would have instantly high temperatures. Not having a functional pump because it is not plugged in correctly would be a problem too.
Check the Corsair link application to see if there are adequate pump RPMs being reported, or look in the bios at the primary CPU or Pump header sensor data to see that one or the other, depending on your board and where you plugged the pump in at, is reporting RPMs indicating a running pump. You can also put your finger on the top of the pump to see if you can feel any vibration indicating the pump is running at all.
Furthermore, if you are not running version 26.6 of Prime95 then you are testing with AVX instruction sets, and that's not recommended. If that's the case you need to download version 26.6 and run ONLY the Small FFT option. Blend is worthless for thermal or stability testing as it does not provide a steady state load and primarily tests memory. Large FFT is not the correct method either. Small FFT only.
http://windows-downloads-center.blogspot.com/2011/04/prime95-266.html
If everything on your pump is correctly connected AND you are running the correct version of Prime, and the fans on your radiator seem to be working correctly then it's possible you have a faulty pump. Checking to see what the pump or CPU header RPMs say, or whether you can feel the vibration of the pump running when you touch it with your finger would be the best place to start, PRIOR to running any stress tests.
It may simply be due to running the wrong version of Prime though.
This pretty well sums things up and is equally relevant whether working with an Intel or an AMD system.
Using x264 encoding or AVX / AVX2 / FMA3 apps won't work as a unilateral metric for thermal testing.
(1) A steady-state workload gives steady-state temperatures; encoding does not.
(2) Simplicity in methodology; most users would find encoding apps unfamiliar and cumbersome to accomplish a simple task.
(3) Most users such as gamers never run any apps which use AVX / FMA, so adaptive or manual voltage aside, it makes no sense to downgrade your overclock to accommodate those loads and temps.
(4) Standardization; Prime95 has been around since 1996; many users are familiar with it.
For the minority of users who routinely run AVX / FMA apps, then P95 v28.5 can be useful tweaking BIOS for thermal and stability testing.
regardless of architecture. P95 v26.6 works equally well across all platforms. Steady-state is the key. How can anyone extrapolate accurate Core temperatures from workloads that fluctuate like a bad day on the Stock Market?
I'm aware of 5 utilities with steady-state workloads. In order of load level they are:
(1) P95 v26.6 - Small FFT's
(2) HeavyLoad - Stress CPU
(3) FurMark - CPU Burner
(4) Intel Processor Diagnostic Tool - CPU Load
(5) AIDA64 - Tools - System Stability Test - Stress CPU
AIDA64's Stress CPU fails to load any overclocked / overvolted CPU to get anywhere TDP, and is therefore useless, except for giving naive users a sense of false security because their temps are so low.
HeavyLoad is the closest alternative. Temps and watts are within 3% of Small FFT's.