Definite Integral Definition: ∫[a,b] f(x)dx = lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx, where Δx = (b-a)/n and xᵢ* is any sample point in the i-th subinterval [xᵢ₋₁, xᵢ].
- Identify the interval [a, b] from the limits of the sum
- Identify the function f(x) from the summand
- Recognize that Δx represents the width of subintervals
- Replace xᵢ* with x and the sum with the integral symbol
The problem states that we want the integral on the interval [0, 2]
So a = 0 and b = 2
In the Riemann sum ∑[i=1 to n] (3xᵢ* + 2)Δx, the function evaluated at the sample point is (3xᵢ* + 2)
Replacing the sample point xᵢ* with the variable x, we get f(x) = 3x + 2
Δx represents the width of each subinterval
For n subintervals on [a, b]: Δx = (b - a)/n = (2 - 0)/n = 2/n
By the definition of the definite integral:
lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx = ∫[a,b] f(x)dx
Therefore: lim[n→∞] ∑[i=1 to n] (3xᵢ* + 2)Δx = ∫[0,2] (3x + 2)dx
∫[0,2] (3x + 2)dx = [3x²/2 + 2x][0,2]
= (3(4)/2 + 4) - (0 + 0) = 6 + 4 = 10
The limit lim[n→∞] ∑[i=1 to n] (3xᵢ* + 2)Δx equals the definite integral ∫[0,2] (3x + 2)dx.
• Riemann Sum Definition: lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx = ∫[a,b] f(x)dx
• Subinterval Width: Δx = (b - a)/n
• Sample Point: xᵢ* is any point in the i-th subinterval [xᵢ₋₁, xᵢ]
Right Riemann Sum: When xᵢ* is the right endpoint of the i-th subinterval, xᵢ* = a + iΔx = xᵢ. The sum becomes ∑[i=1 to n] f(xᵢ)Δx.
The problem specifies the interval [1, 4]
So a = 1 and b = 4
The summand is (xᵢ*)², so the function is f(x) = x²
Since xᵢ* is the right endpoint of each subinterval:
xᵢ* = xᵢ = a + iΔx = 1 + i·(4-1)/n = 1 + 3i/n
By the definition of the definite integral:
lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx = ∫[a,b] f(x)dx
Therefore: lim[n→∞] ∑[i=1 to n] (xᵢ*)² Δx = ∫[1,4] x² dx
∫[1,4] x² dx = [x³/3][1,4]
= (4³/3) - (1³/3) = 64/3 - 1/3 = 63/3 = 21
The limit lim[n→∞] ∑[i=1 to n] (xᵢ*)² Δx equals the definite integral ∫[1,4] x² dx.
• Right Endpoint Formula: xᵢ = a + iΔx
• Definite Integral Definition: Limit of Riemann sums
• Power Rule Integration: ∫x² dx = x³/3
Exponential Function Integration: The definite integral ∫[a,b] eˣ dx = eᵇ - eᵃ. This can be proven using the definition of the definite integral as a limit of Riemann sums.
For ∫[0,1] eˣ dx, we have a = 0, b = 1
Δx = (b - a)/n = (1 - 0)/n = 1/n
Using right endpoints: xᵢ = a + iΔx = 0 + i(1/n) = i/n
So: ∫[0,1] eˣ dx = lim[n→∞] ∑[i=1 to n] e^(i/n) · (1/n)
∑[i=1 to n] e^(i/n) · (1/n) = (1/n) ∑[i=1 to n] e^(i/n)
= (1/n) ∑[i=1 to n] (e^(1/n))^i
The sum ∑[i=1 to n] (e^(1/n))^i is a geometric series with:
First term: a = e^(1/n)
Common ratio: r = e^(1/n)
Number of terms: n
Sum = e^(1/n) · [1 - (e^(1/n))^n] / [1 - e^(1/n)]
= e^(1/n) · [1 - e] / [1 - e^(1/n)]
lim[n→∞] (1/n) · e^(1/n) · [1 - e] / [1 - e^(1/n)]
= (1 - e) · lim[n→∞] e^(1/n) / [n(1 - e^(1/n))]
As n → ∞, e^(1/n) → 1
We need to evaluate lim[n→∞] 1/[n(1 - e^(1/n))]
Let u = 1/n, so as n → ∞, u → 0
lim[u→0] 1/[u(1 - e^u)] = lim[u→0] 1/[u(1 - e^u)]
Using L'Hôpital's rule: lim[u→0] 1/(1 - e^u - ue^u) = 1/(0 - 1) = -1
Actually, let's use: lim[u→0] u/(1 - e^u) = lim[u→0] 1/(-e^u) = -1
So lim[u→0] 1/[u(1 - e^u)] = -1
Therefore: (1 - e) · (-1) = e - 1
Using the fundamental theorem: ∫[0,1] eˣ dx = [eˣ][0,1] = e¹ - e⁰ = e - 1 ✓
The definite integral ∫[0,1] eˣ dx equals e - 1.
• Riemann Sum Definition: Converting integral to limit of sums
• Geometric Series: Sum of r^i from i=1 to n
• L'Hôpital's Rule: For indeterminate forms
• Exponential Integration: ∫eˣ dx = eˣ
Riemann Sum: An approximation of the area under a curve using rectangles. The sum ∑[i=1 to n] f(xᵢ*)Δx where xᵢ* is a sample point in the i-th subinterval.
Definite Integral: The exact area under the curve, defined as the limit of Riemann sums as n approaches infinity.
Sample Point: Any point xᵢ* chosen from the i-th subinterval [xᵢ₋₁, xᵢ] for function evaluation.
- Partition the interval: Divide [a,b] into n equal subintervals
- Calculate Δx: Find the width of each subinterval: Δx = (b-a)/n
- Choose sample points: Select xᵢ* in each subinterval (left, right, midpoint, or arbitrary)
- Form the Riemann sum: Calculate ∑[i=1 to n] f(xᵢ*)Δx
- Take the limit: Find lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx
- Verify with FTC: Compare with antiderivative method when possible
• Definite Integral Definition: ∫[a,b] f(x)dx = lim[n→∞] ∑[i=1 to n] f(xᵢ*)Δx
• Subinterval Width: Δx = (b-a)/n
• Left Endpoint: xᵢ* = a + (i-1)Δx
• Right Endpoint: xᵢ* = a + iΔx
• Midpoint: xᵢ* = a + (i-1/2)Δx
Riemann Sum Foundation: The connection between discrete approximations (finite sums) and continuous quantities (definite integrals) through the limit process.
Limit Process: The fundamental idea that as the number of subintervals increases indefinitely, the Riemann sum approaches the exact area under the curve.
Sample Point Flexibility: The specific choice of sample points within subintervals becomes irrelevant in the limit as n approaches infinity.
- Identify the correspondence: Match the Riemann sum components to integral components
- Recognize the pattern: Identify the function f(x) from the summand
- Determine the interval: Establish the bounds of integration [a,b]
- Express the limit: Write the Riemann sum in terms of n and Δx
- Take the limit: Evaluate lim[n→∞] to find the definite integral
- Verify results: Compare with the Fundamental Theorem when possible
• The definite integral is defined as the limit of Riemann sums, not computed directly from this definition
• The Fundamental Theorem of Calculus provides a much more practical method for evaluating definite integrals
• Riemann sums are essential for understanding the theoretical foundation of integration
• Different sample point choices lead to the same limit, confirming the integral's uniqueness
Norm of Partition: The length of the largest subinterval. For uniform partitions, it's simply Δx = (b-a)/n.
Refinement: A partition where every point of the original partition is included, making the norm smaller.
Integrability: A function f is integrable on [a,b] if the limit of Riemann sums exists and is the same regardless of sample point choice.
- Uniform vs Non-uniform Partitions: Equal vs unequal subinterval widths
- Special Sample Points: Left, right, midpoint, or arbitrary points within subintervals
- Geometric Series Applications: For exponential functions in Riemann sums
- L'Hôpital's Rule: For evaluating difficult limits in the conversion process
• Numerical integration methods (trapezoidal rule, Simpson's rule)
• Probability theory (expectation values as integrals)
• Physics (work done by variable forces)
• Engineering (moment calculations, center of mass)
Questions & Answers
Question: Why do we need the limit process? Can't we just use the Fundamental Theorem directly?
Answer: Great question! Here's why both are important:
- Theoretical Foundation: The limit of Riemann sums defines what a definite integral actually means. Without this definition, the Fundamental Theorem would have no meaning.
- Practical Computation: The Fundamental Theorem provides a much more efficient method for calculating definite integrals, so we typically use it for computations.
- Understanding: Riemann sums help us understand what integration represents geometrically (area under the curve).
- Proofs: Many important theorems in calculus rely on the Riemann sum definition.
Think of it this way: Riemann sums tell us what integration is, and the Fundamental Theorem tells us how to calculate it efficiently.
In Exercise 3, we showed how to evaluate ∫[0,1] eˣ dx using the definition, but it's much easier to use the antiderivative: [eˣ][0,1] = e - 1.
Question: What's the difference between left, right, and midpoint Riemann sums, and why does it not matter in the limit?
Answer: The differences are in how we choose the sample point xᵢ* in each subinterval [xᵢ₋₁, xᵢ]:
- Left Riemann Sum: xᵢ* = xᵢ₋₁ (left endpoint)
- Right Riemann Sum: xᵢ* = xᵢ (right endpoint)
- Midpoint Rule: xᵢ* = (xᵢ₋₁ + xᵢ)/2 (midpoint)
Why it doesn't matter in the limit: As n → ∞, Δx → 0, so the difference between sample points within each subinterval approaches zero. For continuous functions, f(x) changes very little over tiny intervals.
The error between different choices is roughly proportional to Δx, so as Δx → 0, the error → 0. This is why all valid Riemann sums converge to the same limit.
However, for finite n, different choices can give significantly different approximations!
Question: How do I convert a definite integral back to a Riemann sum?
Answer: To convert ∫[a,b] f(x)dx to a Riemann sum:
- Identify [a,b]: These become the limits of the definite integral
- Identify f(x): This is the function to be evaluated at sample points
- Set up partition: Δx = (b-a)/n
- Choose sample points: Typically right endpoints: xᵢ = a + iΔx
- Form the sum: ∑[i=1 to n] f(xᵢ)Δx
- Take the limit: lim[n→∞] ∑[i=1 to n] f(xᵢ)Δx
For example, ∫[2,5] x² dx becomes:
Δx = (5-2)/n = 3/n
xᵢ = 2 + i(3/n) = 2 + 3i/n
∫[2,5] x² dx = lim[n→∞] ∑[i=1 to n] (2 + 3i/n)² · (3/n)
The choice of sample points (left, right, midpoint) doesn't affect the final answer in the limit.
Question: What happens if the function is not continuous? Does the Riemann sum still work?
Answer: This is a sophisticated question! The situation is more nuanced:
- Continuous functions: Are always Riemann integrable
- Bounded functions with finite discontinuities: Are Riemann integrable
- Functions with too many discontinuities: May not be Riemann integrable
For functions with finitely many jump discontinuities, the Riemann sum limit still exists and equals the definite integral. The discontinuities have "zero width" and don't affect the area.
However, for functions like Dirichlet's function (1 if rational, 0 if irrational), the Riemann sum limit doesn't exist because the choice of sample points dramatically affects the sum.
In Calculus 1, we typically work with continuous functions where Riemann integration always works.
Question: How can I visualize the connection between Riemann sums and definite integrals?
Answer: Here's an effective visualization approach:
- Start with rectangles: Draw rectangles under the curve using any Riemann sum method
- Count the rectangles: Begin with n=4, then increase to n=8, n=16, etc.
- Observe the pattern: Notice how the rectangles better approximate the area as n increases
- Imagine the limit: As rectangles become infinitely thin (Δx→0), they perfectly conform to the curve
Think of it like zooming in: at first you see distinct rectangles, but as you zoom in infinitely, the tops of the rectangles become the curve itself.
The Riemann sum is like a "pixelated" version of the area, and the definite integral is the "high-resolution" version. As pixel size (Δx) approaches zero, the pixelated image approaches the smooth curve.
The limit process transforms the discrete sum of rectangle areas into the continuous area under the curve.